Untitled

HEALTHCARE TECHNOLOGIES SERIES 42
Smart Health
Technologies for the
COVID-19 Pandemic
IET Book Series on e–Health Technologies
Book Series Editor: Professor Joel J.P.C. Rodrigues, College of Computer Science and
Technology, China University of Petroleum (East China), Qingdao, China; Senac Faculty of
Ceará, Fortaleza-CE, Brazil and Instituto de Telecomunicações, Portugal
Book Series Advisor: Professor Pranjal Chandra, School of Biochemical Engineering, Indian
Institute of Technology (BHU), Varanasi, India
While the demographic shifts in populations display significant socio-economic challenges, they
trigger opportunities for innovations in e-Health, m-Health, precision and personalized
medicine, robotics, sensing, the Internet of things, cloud computing, big data, software defined
networks, and network function virtualization. Their integration is however associated with
many technological, ethical, legal, social, and security issues. This book series aims to
disseminate recent advances for e-health technologies to improve healthcare and people’s
wellbeing.
Could you be our next author?
Topics considered include intelligent e-Health systems, electronic health records, ICT-enabled
personal health systems, mobile and cloud computing for e-Health, health monitoring,
precision and personalized health, robotics for e-Health, security and privacy in e-Health,
ambient assisted living, telemedicine, big data and IoT for e-Health, and more.
Proposals for coherently integrated international multi-authored edited or co-authored

handbooks and research monographs will be considered for this book series. Each proposal will
be reviewed by the book Series Editor with additional external reviews from independent
reviewers.
To download our proposal form or find out more information about publishing with us, please
visit https://www.theiet.org/publishing/publishing-with-iet-books/.
Please email your completed book proposal for the IET Book Series on e-Health Technologies
to: Amber Thomas at athomas@theiet.org or author_support@theiet.org.
Smart Health
Technologies for the
COVID-19 Pandemic
Internet of medical things perspectives
Edited by
Chinmay Chakraborty and Joel J.P.C. Rodrigues
The Institution of Engineering and Technology

Published by The Institution of Engineering and Technology, London, United Kingdom
The Institution of Engineering and Technology is registered as a Charity in England &
Wales (no. 211014) and Scotland (no. SC038698).
† The Institution of Engineering and Technology 2022
First published 2022
This publication is copyright under the Berne Convention and the Universal Copyright
Convention. All rights reserved. Apart from any fair dealing for the purposes of research
or private study, or criticism or review, as permitted under the Copyright, Designs and
Patents Act 1988, this publication may be reproduced, stored or transmitted, in any
form or by any means, only with the prior permission in writing of the publishers, or in
the case of reprographic reproduction in accordance with the terms of licences issued
by the Copyright Licensing Agency. Enquiries concerning reproduction outside those
terms should be sent to the publisher at the undermentioned address:
The Institution of Engineering and Technology

Michael Faraday House
Six Hills Way, Stevenage
Herts, SG1 2AY, United Kingdom
www.theiet.org
While the authors and publisher believe that the information and guidance given in this
work are correct, all parties must rely upon their own skill and judgement when making
use of them. Neither the authors nor publisher assumes any liability to anyone for any
loss or damage caused by any error or omission in the work, whether such an error or
omission is the result of negligence or any other cause. Any and all such liability is
disclaimed.
The moral rights of the authors to be identified as authors of this work have been
asserted by them in accordance with the Copyright, Designs and Patents Act 1988.
British Library Cataloguing in Publication Data

A catalogue record for this product is available from the British Library
ISBN 978-1-83953-518-5 (hardback)

ISBN 978-1-83953-519-2 (PDF)
Typeset in India by MPS Limited

Printed in the UK by CPI Group (UK) Ltd, Croydon
Contents
About the editors xv

Preface xvii
1 Internet of Things (IoT) and blockchain-based solutions to confront

COVID-19 pandemic 1
Abu Hasnat Md Rhydwan, Md Mashrur Sakib Choyon,
A.S.M. Mehedi Hasan Sad, Kazi Ahmed Asif Fuad, Kawshik Shikder,
Chowdhury Akram Hossain and M. Shamim Kaiser
1.1 Introduction 2
1.2 Internet of Things (IoT) and blockchain overview 3
1.2.1 Internet of Things 4
1.2.2 Blockchain 6
1.3 IoT technologies to confront COVID-19 9
1.3.1 Health monitoring systems 10
1.3.2 Tracking and detecting possible patients 12
1.3.3 Disinfecting area 13
1.3.4 Telemedicine 14
1.3.5 Logistics delivery 14
1.4 Blockchain technologies to confront COVID-19 14
1.4.1 Contact tracing 15
1.4.2 Database security 16
1.4.3 Information sharing 16
1.4.4 Prevention of data fabrication 17
1.4.5 Internet of Medical Things 18
1.5 Challenges, solutions, and deliverables 18
1.5.1 Challenges of IoT and blockchain technology 18
1.5.2 Possible solutions and deliverables 19
1.6 Key findings and discussion 20
1.7 Conclusion and future scopes 21
References 22
2 Application of big data and computational intelligence in fighting

COVID-19 epidemic 33
Joseph Bamidele Awotunde, Chinmay Chakraborty and
Gbemisola Janet Ajamu
2.1 Introduction 34
vi Smart health technologies for the COVID-19 pandemic
2.2 Applicability of computational intelligence in combating

COVID-19 pandemic 36
2.3 Big data and analytics in battling COVID-19 outbreak 40
2.4 The limitations of using big data and computational intelligence
to fight the COVID-19 pandemic 44
2.5 The practical case of using computational intelligence in fighting
COVID-19 pandemic 48
2.5.1 Confusion matrix 49
2.5.2 ROC curves 50
2.5.3 Precision-recall curve 50
2.6 Conclusion 51
References 51
3 Cloud-based IoMT for early COVID-19 diagnosis and monitoring 61

G. Boopathi Raja, T. Sathya, V. Gowrishankar and M. Parimala Devi
3.1 Introduction 62
3.2 Overview about COVID-19 treatments 63
3.2.1 Symptoms 63
3.2.2 Methodologies in COVID-19 diagnosis 63
3.2.3 Treatment approaches 64
3.2.4 Available vaccine 65
3.2.5 COVID-19 timeline 66
3.3 Related work 66
3.3.1 Lightweight block encryption–based secure health
monitoring system for data management 66
3.3.2 Smart diagnostic/therapeutic framework for COVID-19
patients 69
3.3.3 IoT-based framework for collecting real-time symptom
data using machine learning algorithms 70
3.4 Proposed methodology 72
3.4.1 Architecture of proposed IoT framework 73
3.4.2 Data acquisition using wearables devices 76
3.5 Implementation of proposed framework 76
3.6 Results and discussion 78
References 82
4 Assessment analysis of COVID-19 on the global economics

and trades 85
Hemanta Kumar Bhuyan and Chinmay Chakraborty
4.1 Introduction 86
4.2 Backgrounds 87
4.3 Social impacts on finance 88
Contents vii
4.4 Framework for the international financial system,

bionetworks, and maintainability on pandemic 89
4.4.1 Assessment strategy constructions to fight COVID-19 89
4.4.2 Macro-finance impacts 89
4.4.3 Econometric effects: consumer preferences 90
4.4.4 Nonpositive impacts of COVID-19 92
4.4.5 Impact of international commercial trading 94
4.4.6 COVID-19’s effect on the aviation industry 94
4.4.7 Significant collision on the travel sector 96
4.4.8 Significant reduction in primary energy usage 98
4.4.9 Record decrease in CO2 emissions 98
4.4.10 Rise in digitalization 99
4.5 The role of circular economy 100
4.5.1 The circular economy for slowing the onset of climate
collapse 101
4.5.2 Social finance system 102
4.5.3 Hurdles to CE for context of COVID-19 103
4.6 Chances financial support after COVID-19 104
4.6.1 Several solutions to manage hospital medical and
general waste 104
4.6.2 Facilities for CE in communication sector 106
4.6.3 Use digitalization after COVID-19 107
4.7 Conclusions 108
References 110
5 Early diagnosis and remote monitoring using cloud-based

IoMT for COVID-19 115
Madhura S. Mulimani, Shridhar Allagi and Rashmi R. Rachh
5.1 Introduction 116
5.2 Detection techniques 117
5.3 Internet of Medical Things 119
5.4 IoMT devices for the identification of COVID-19 symptoms
and remote monitoring 121
5.4.1 Wearables 122
5.4.2 Smartphone applications 125
5.5 Early diagnosis of COVID-19 and remote monitoring procedures 125
5.6 Machine learning and deep learning in COVID-19 diagnosis 127
5.7 Related works 129
5.8 Experimental case study 129
5.8.1 Dataset description 129
5.8.2 Methodology 130
5.8.3 Training 133
5.8.4 Experimental setup and results 134
5.9 Measures for monitoring and tracking COVID-19 135
viii Smart health technologies for the COVID-19 pandemic
5.10 Limitations of using IoMT devices 136

5.11 Conclusion and future scope 137
References 137
6 Blockchain technology for secure COVID-19 pandemic

data handling 141
Agbotiname Lucky Imoize, Daisy Osarugue Irabor,
Peter Anuoluwapo Gbadega and Chinmay Chakraborty
6.2 Recent developments in blockchain technology 144
6.2.1 Healthcare data systems 147
6.2.2 Healthcare data exchanges 149
6.2.3 Healthcare administration 149
6.2.4 Pharmaceuticals 150
6.3 Potential benefits of blockchain technology in data handling 151
6.3.1 Better exchange of healthcare data records 152
6.3.2 Validating trust in medical research and supplies 152
6.3.3 Validating correct billing management 153
6.3.4 Internet of Things (IoT) in healthcare 153
6.3.5 Optimized privacy and data security 154
6.4 Key challenges of blockchain technology in data handling 154
6.4.1 Security 155
6.4.2 Speed 155
6.4.3 Interoperability 155
6.4.4 Stringent data protection regulation 155
6.4.5 Scalability 156
6.4.6 Privacy 156
6.5 Prospects of blockchain technology 157
6.6 Research on blockchain technology in COVID-19 healthcare 160
6.7 Real-time analysis of COVID-19 pandemic data 163
6.7.1 The susceptible recovered infectious (SIR) model 163
6.7.2 Standard logistic regression model 164
6.7.3 Time-to-event analytics model 164
6.7.4 Results of major real-time analysis 165
6.8 Recommendations and future directions 170
Acknowledgments 173
References 173
7 Social distancing technologies for COVID-19 181

Aumnat Tongkaw
7.2 Methodology 182
7.3 Social distancing technologies for education 182
Contents ix
7.3.1 Learning management system 183

7.3.2 Social networking and conference software for education 186
7.4 Social distancing technology in healthcare 187
7.4.1 Wearable technology 187
7.4.2 Screening system 188
7.4.3 Queue systems 188
7.4.4 Payment system 189
7.4.5 Social distancing notified people in public 191
7.5 Social distancing technology in manufacturing 193
7.5.1 Checking the distance using wearable device 193
7.5.2 Distance monitoring using Wi-Fi 194
7.5.3 Distance monitoring using video analytics 194
7.5.4 Social distancing by replacing some work with a robot 195
7.6 Social-distancing technologies for supporting everyday life 195
7.6.1 Technologies support working at home 196
7.6.2 Applications support work from home (WFH) service 196
7.6.3 Conferencing application 200
7.7 Social distancing and smart city 202
7.7.1 AI and big data 202
7.7.2 Implementation and usability 202
7.7.3 Privacy and security 203
7.7.4 Policy and legislation 203
7.8 Conclusion and future works 203
References 205
8 Social health protection in touristic destinations during COVID-19 209

Zaklina Spalevic, Aleksandra Stojnev Ilic and Milos Ilic
8.2 Related work 212
8.3 Proposal of software solution for health protection 214
8.3.1 System architecture 215
8.3.2 Healthcare service 217
8.3.3 Tourist service 218
8.3.4 Local government service 219
8.3.5 Border control 220
8.4 Data protection 220
References 223
9 Analysis of Artificial Intelligence and Internet of Things in

biomedical imaging and sequential data for COVID-19 227
Sinthia Roy Banerjee, Saurav Mallik, Tapas Si, Arijit Banerjee,
Shan Jiang and Sudip Podder
x Smart health technologies for the COVID-19 pandemic
9.2 Definition of biomedical keywords 230

9.2.1 Microarray and RNA-seq data 230
9.2.2 De novo mutation 231
9.2.3 ChiP-seq data 231
9.2.4 Biomedical imaging 231
9.3 Categories of computational algorithms in biomedical data 232
9.3.1 Biomedical data analysis 232
9.3.2 Array-based data analysis 233
9.3.3 Hybrid data analysis 235
9.4 Different techniques for diagnosis using biomedical imaging 235
9.4.1 Brain 235
9.4.2 Breast 236
9.4.3 Kidney 236
9.4.4 Ovary 237
9.4.5 Skin cancer 237
9.4.6 Soft tissue sarcoma 238
9.5 Comparative review of computational algorithms 238
9.6 Role of CT in COVID-19 pandemic 238
9.7 Advent of smart technologies during COVID-19 251
9.7.1 Building ML models to diagnose COVID-19 253
9.7.2 Impact of IoT in healthcare 253
9.8 Conclusion 254
References 255
10 Review of medical imaging with machine learning and deep

learning-based approaches for COVID-19 261
Swapnil Singh, Vidhi Vazirani and Deepa Krishnan
10.2 Literature review 264
10.2.1 Reviewed work 264
10.3 Comparative analysis of existing work 279
10.4 Research gaps 289
10.4.1 Unavailability of large datasets 289
10.4.2 Imbalanced datasets 289
10.4.3 Multiple image sources 290
10.5 Conclusion 290
References 291
11 Machine-based drug design to inhibit SARS-CoV-2 virus 295

T. Lurthu Pushparaj, E. Francy Irudaya Rani,
E. Fantin Irudaya Raj and M. Appadurai
11.2 What is SARS-coronavirus-2? 298
Contents xi
11.3 Mechanism of SARS-coronavirus-2 infection in human 299

11.4 How SARS-coronavirus-2 multiplies? 300
11.5 Human antibody generation and role of vaccine 302
11.5.1 Immediate action of human antibody 302
11.5.2 Role of synthetic vaccine on COVID-19 302
11.6 Real-time COVID-19 identification test (RT-PCR) 303
11.6.1 Limitations of RT-PCR tool 304
11.7 Discussion on in silico methods in COVID-19 drug research 305
11.7.1 In silico–assisted anchoring site analysis 305
11.7.2 Machine-assisted designing and evaluation of
COVID-19 drug 307
11.8 Machine-integrated advanced techniques for COVID-19 318
11.8.1 Computerized tomography in COVID-19 detection 318
11.8.2 Advanced MRI for COVID-19 treatment 319
11.9 Summary 322
11.10.1 Future scope 324
References 325
12 Stress detection for cognitive rehabilitation in

COVID-19 scenario 331
Ahona Ghosh, Sima Das and Sriparna Saha
12.2 Related works 333
12.3 Proposed framework 335
12.3.1 Introduction to EEG 340
12.3.2 Feature extraction using DWT 341
12.3.3 Feature selection using principal component analysis 342
12.3.4 Classification using support vector machine 343
12.4 Experimental outcomes and discussions 344
12.4.1 Dataset preparation 344
12.4.2 sLORETA-based activated brain region selection 345
12.4.3 Discrete wavelet transform–based feature extraction
outcome 345
12.4.4 Principal component analysis–based dimensionality
reduction outcome 346
12.4.5 Support vector machine–based classification outcome 346
12.4.6 Performance metrics 348
12.4.7 Performance evaluation 348
12.4.8 Statistical significance using t-test 350
Acknowledgment 351
References 352
xii Smart health technologies for the COVID-19 pandemic
13 Arduino-based robot for purification of COVID-19 using far

UVC light 359
C.N. Sujatha, B. Sri Charan and K. Himabindu
13.1.1 Arduino 360
13.1.2 Far-UVC lamp 363
13.2 Literature survey 365
13.2.1 Improvements and requirements 372
13.3 Working of the proposed robot 376
13.3.1 Value proposition 377
13.4 Results and discussions 378
References 381
14 Effect of COVID-19 pandemic on waste management system

and infection control 385
Ramkrishna Mondal and Chinmay Chakrabarty
14.2 Socioeconomic and environmental impact 387
14.3 Impact of waste generation 388
14.4 Impacts on waste management 390
14.4.1 Waste management adjustments 392
14.5 Challenges in handling waste 393
14.6 Rethinking effective waste management 394
14.6.1 Policy, regulatory, and guidelines 395
14.6.2 Handling of infectious waste 395
14.6.3 Suitable disposal methods 396
14.6.4 Information, education, and communication 396
14.6.5 Data management and learning 396
14.6.6 Monitoring of segregation 396
14.6.7 Basic principles for managing waste 396
14.6.8 Fund raising and national and international collaboration 397
References 398
15 Natural adjunctive therapies options other than COVID-19

antiviral therapies 405
Betul Ozdemir and Zeliha Selamoglu
15.2 Immune system and inflammatory responds 407
15.3 Proinflammatory cytokines 408
15.4 Immunomodulators and adjunctive therapies 409
15.4.1 Phenolic compounds 409
Contents xiii
15.4.2 Melatonin 412

15.4.3 Zinc 413
15.4.4 Ascorbic acid (vitamin C) 413
15.4.5 Vitamin D 413
15.4.6 Vitamin E 414
15.4.7 Selenium 414
15.4.8 Omega-3 fatty acids 414
15.5 Dietary ingredients in immunity 415
15.6 Conclusion and future scope for natural antiviral
therapies against COVID-19 415
References 415
16 Risk assessment and spread of COVID-19 421

Challa Sri Gouri, D. Ajitha, Nikhil Mulaguru and Goteti Rithika
16.2 Technology and epidemics 422
16.2.1 Healthcare 425
16.2.2 Education 425
16.2.3 Work 425
16.2.4 Others 425
16.3 Prediction techniques 426
16.4 General methods followed for risk assessment 427
16.4.1 What-if analysis 428
16.4.2 Fault-tree analysis 429
16.4.3 Guidelines issued by World Health Organization 430
16.5 Prevention and management of epidemics 431
16.5.1 Strategies proposed 432
16.5.2 Sentimental analysis using machine learning 433
16.6 Protecting the living beings from the impact of epidemics 434
16.6.1 Impact of COVID-19 on agriculture sector 434
16.6.2 Impact of COVID-19 on economy 434
16.6.3 Impact of COVID-19 on educational sector 435
16.7 Our contribution 435
16.7.1 Proposed method and its working 435
16.7.2 Components required 436
16.7.3 Software required and simulation 437
References 439
Index 443
This page intentionally left blank
About the editors
Chinmay Chakraborty, Birla Institute of Technology, India
Chinmay Chakraborty has published more than 70 conference presentations, jour-

nal papers, book chapters and books. He has served on the Editorial Boards of more
than ten journals, including Future Internet Journal (Wiley), Internet Technology
Letters (Springer), and Advances in Smart Healthcare Technologies (CRC Press)
and the organizing committees of numerous IEEE international conferences.
Joel J.P.C. Rodrigues, College of Computer Science and Technology, China

University of Petroleum (East China), Qingdao, China; Senac Faculty of Ceará,
Fortaleza-CE, Brazil and Instituto de Telecomunicações, Portugal
Joel J.P.C. Rodrigues is an Highly Cited Researcher has published over 1,000
papers in refereed international journals and conferences, 3 books, 2 patents, and 1
ITU-T Recommendation. He is the editor-in-chief of the International Journal of
E-Health and Medical Communications and an editorial board member of several
high-reputed journals. He has chaired many international conferences, including
IEEE ICC, IEEE GLOBECOM, IEEE HEALTHCOM, and IEEE LatinCom. He
has received several Outstanding Leadership and Outstanding Service Awards by
IEEE Communications Society. He is a member of the Internet Society, a senior
member ACM, and a Fellow of AAIA and IEEE.
Preface
Intelligent health technologies are in the revolution phase due to recent develop-
ments in sensor technology, wearables, edge computing, the Internet of Things, and
enabling wireless communication protocols, creating value-based care delivery in
terms of outcome monitoring, therapeutic and preventive interventions. This book
focuses the COVID-19 monitoring and analysis using intelligent e-healthcare sys-
tems. The innovative solutions might include (but are not limited to) detection and
treatment methods for COVID-19, strategies to find suspected patients, contact
tracing of infected patients, intelligent strategies to predict future pandemic or
epidemic outbreaks, informatics-based policies to contain such mass infectious
outbreaks, ensuring the safety of patients, careers, and the public, information
exchange, and knowledge sharing.
This book will include reviews and original works on the COVID-19 in terms
of e-healthcare, telemedicine, other medical technology, life support systems, fast
detection, and diagnoses, developed technologies and innovative solutions, bioin-
formatics, datasets, apps for diagnosis, solutions for monitoring, and control of the
spread of COVID-19, among various other topics. The book focuses on recent
advances in the Internet of Medical Things in smart advanced healthcare technol-
ogies. Intelligent advanced healthcare informatics plays a vital role in this crucial
time. The objective of the book is to present innovative technological solutions
utilizing informatics to deal with various COVID-19 outbreak issues. This book
covers comprehensive studies from healthcare informatics, biomedical engineering,
artificial intelligence, big data, blockchain, and data mining with a prime focus on
the COVID-19 pandemic.
At last, we would like to extend our sincere thanks to authors from industry,
academia, and policy expertise to complete this work for aspiring researchers in
this domain. We are confident that this book would play a key role in providing
readers a comprehensive view of COVID-19 health informatics and developments
around it and can be used as a learning resource for various examinations, which
deal with cutting-edge technologies.
Book organisation
The book consists of 16 chapters in the field of COVID-19 healthcare systems. A
summary of each chapter is presented in the following.
xviii Smart health technologies for the COVID-19 pandemic
Chapter 1: Internet of Things (IoT) and blockchain-based

solutions to confront COVID-19 pandemic
In this demanding period, the applications of the latest technologies to prevent the
spread of the virus are critical. Among various technologies, the Internet of Things
(IoT) and blockchain are being used in several solutions starting from contact tra-
cing to forecasting. Blockchain-based solutions are also being used during the
ongoing pandemic in various aspects for secure contact tracing, data handling, and
preventing data fabrication. In this chapter, IoT and blockchain technology are
discussed briefly while describing their core elements. Then, the latest solutions
were presented based on these technologies in different aspects of COVID-19
pandemic prevention. These solutions mainly focus on using these technologies for
remote patient monitoring, secure data handling, and telemedicine.
Chapter 2: Application of big data and computational

intelligence in fighting COVID-19 epidemic
Big data is driving the digital revolution in an increasingly knowledge-driven,
healthcare-innovation-driven, and connected society. The combination of computa-
tional intelligence (CI) and big data analytics (BDA) has developed methods that
make accessing and processing vast amounts of data easier and less demanding on
human expert. This chapter reviews the applicability and importance of big data and
CI methods to data produced from the countless ubiquitously connected healthcare
devices that produced entrenched and distributed information handling capabilities in
fighting the COVID-19 outbreak. The use of CI in BDA has resulted in a knowledge-
based system that transformed big data into big knowledge with new approaches and
visions to provide people with better understanding and information-driven results.
Chapter 3: Cloud-based IoMT for early COVID-19

diagnosis and monitoring
This chapter introduces an effective scheme for the diagnosis and tracking of patients
with COVID-19 based on symptoms. The concept behind the proposed approach is to
use an IoT-based system to handle the real-time symptom data from patients to
diagnose coronavirus cases early. The proposed framework must learn automatically
about the origin of the virus by monitoring and analyzing necessary data. The health
status of patients is determined in this model by predicting critical situations and
examining physiological data received from smart medical IoT devices, ensuring that
patients’ personal information is kept private. Based on current advancements, the
suggested model is thought to be suitable for delivering an appropriate remote patient
monitoring model with accurate data in cloud-based IoT systems.
Chapter 4: Assessment analysis of COVID-19 on the global

economics and trades
In light of the epidemic’s context, this chapter gives a critical evaluation of the
inventory of the pandemic’s bad and positive influences in a different section. This
Preface xix
advocates for a complete overhaul of the global economic development paradigm,

which is based on a linear economy system that leverages profiteering and energy-
guzzling industrial processes. The utilization of publicly accessible data, as found
on Yahoo Finance, the International Monetary Fund, and John Hopkins COVID-
19 map, and regression models were used to carry out the goal of this study. This
study will aid those with government-level decision-making, business-stage stra-
tegic thinking, and capital-market investment to better comprehend the current state
of affairs and use the model for forecasting.
Chapter 5: Early diagnosis and remote monitoring using

cloud-based IoMT for COVID-19
Hence, it is essential to detect the disease at an early stage and ensure that the
transmission of the virus is curbed, in turn, saving the lives of many other people.
With the advancements and developments in the information technology field, it is
possible to diagnose infectious diseases like the current COVID-19 pandemic at an
early stage and give proper treatment to the infected. In this chapter, various IoTs
that are being used to track the patients’ health and provide them the necessary care
and treatment even in remote locations have been discussed. Also, machine learning
and deep learning for early diagnosis and remote monitoring have been discussed. An
experimental case study using COVIDX dataset has been discussed along with the
results. Comprehensive experiments have been carried out with varying CT sizes of
CT images and an average accuracy of above 80% has been achieved.
Chapter 6: Blockchain technology for secure COVID-19

pandemic data handling
The decentralized and immutability features of blockchain have facilitated the
rapid development of the beyond 5G wireless services for health-related data
sharing and processing. However, the adoption of blockchain technology in medi-
cal data handling is still in its infancy, and the need for a rigorous study in this
domain cannot be overemphasized. This chapter first highlights the recent devel-
opments in blockchain technology and its applications in the medical data-handling
space. Further to this, the chapter examines the potential benefits, key challenges,
and prospects in blockchain technology. Additionally, the chapter proposes high-
lighting the recent blockchain-related research efforts within the healthcare
domain, especially as it relates to the dreaded COVID-19 pandemic. The chapter
also discusses how blockchain would be applied to healthcare data-handling prac-
tices to build trust with automated provenance tracking and accountable credential
verification. Finally, practical COVID-19 pandemic data were analyzed and pre-
sented to motivate the chapter further.
Chapter 7: Social-distancing technologies for COVID-19

Social divides have a role in illness prevention. As a result, Thailand is focusing on
adopting different technologies to assist us in managing the gap and support
everyone’s ability to work. Adopting this technology has resulted in it being the
xx Smart health technologies for the COVID-19 pandemic
New Normal in use as a substitute for older systems. This chapter offers a template
analysis of stakeholder interviews conducted with the aid of technology to
accomplish social distancing, which is divided into three categories: educational,
public health, and social distancing, as well as manufacturing employees. A soft-
ware usage pattern has been developed due to data analysis utilizing content ana-
lysis to help create social distancing for job planning and everyday use. It can also
be used to plan future smart cities.
Chapter 8: Social health protection in touristic destinations

during COVID-19
This chapter gives a proposal for a software solution that would significantly
improve the control and registration of infected persons. The straightforward use of
the proposed software solution aims to prevent the spread of the virus by tourists.
The proposed software system should enable better control of COVID-19 positive
individuals, while on the other hand should support the improvement of traveling
conditions and allow the tourism sector to work for people who are not virus-
positive. The system will allow us to check vaccination or home isolation status of
individuals, and based on the results, approve or deny reservations. The edge/cloud
architecture will provide data availability, and at the same time data access control.
Chapter 9: Analysis of Artificial Intelligence and Internet of

Things in biomedical imaging and sequential
data for COVID-19
In this chapter, the authors surveyed various array-based sequence data analyses
and biomedical imaging along with their integrated studies for different tissue-specific
dreadful diseases (such as cancer). The integrated studies of biomedical imaging and
array-based data analysis for the same set of patients (samples) has been discussed that
covered the problem of combinatorial gene signature detection as well as disease
subtype image classifications while specific multimodal data from a well-known data
repository (e.g., TCGA, ICGC) had been provided. This chapter covers the maximum
area of biomedical imaging as well as array-based sequence data analysis along with
the contribution of AI and ML to build a smart healthcare system and provide a new
dimension to the interested biomedical researchers.
Chapter 10: Review of medical imaging with machine

learning and deep learning-based approaches
for COVID-19
Early detection of COVID-19 is inevitable as it helps in seeking early medical
intervention and to minimize community spread. COVID-19 can be diagnosed with
the help of gold standard RT-PCR tests, quick antigen and antibody tests supple-
mented with CT scan, and X-ray imaging. Considering the increase in the number
of cases worldwide, there is a need for economical and viable ways to detect
COVID-19. Medical imaging and analytic tools that use machine learning and deep
learning algorithms can enhance the diagnosis and prediction of COVID-19. There
Preface xxi
are many effective and proven image recognition techniques in deep learning like
Convolutional Neural networks and Transfer learning that can be used to design
very promising applications for COVID-19 detection. They enhance these models
using advanced image segmentation and edge detection techniques. The authors
have been reviewed the impact of the COVID-19 pandemic on the global com-
munity, the need for reliable, quick, and economical ways to detect it.
Chapter 11: Machine-based drug design to inhibit SARS-

CoV-2 virus
Due to the lack of animal and human trials data, the process is still complicating
and causing more death worldwide. The keen analysis of the amino acid sequences
in the coronavirus-19 infection-causing proteins will give very important infor-
mation about the virus transformation and replication cycle. The amino acid
sequences and their active sites provided in the 3D crystalline structure of M-pro
(PDB ID; 6LU7) afford valuable data to the researchers about the type of inhibitors
that corresponds to the severe acute respiratory syndrome (SARS)-coronavirus-2
(SARS-CoV-2) inhibition. Compared to all other deadly viruses like flu, HIV, and
SARS, novel coronavirus SARS-CoV-2 shows superior binding affinity over a
human transmembrane protein christened angiotensin-converting enzyme 2, found
in human lungs. Since the coronavirusid-19 infection spread more vigorously, the
traditional RT-PCR test will need more time for infection confirmation, so
machine-based imaging studies like MRI, CT, and X-ray are needed.
Chapter 12: Stress detection for cognitive rehabilitation in

COVID-19 scenario
This chapter highlights the mobile phone exposure to people increased at least 2-
fold way, so games have been used as stimuli for detecting how our brain becomes
overburdened with increased exposure. After the data acquisition from 14 channels
of an electroencephalogram, the activated regions were identified. Features were
extracted from the most activated ten electrode channels using discrete wavelet
transform. To reduce the dimensions of the feature space for enhancing the per-
formance, principal component analysis was used. The mental state classification
was performed using a support vector machine based on the detected stress. The
proposed system has outperformed the existing ones for its effectiveness and effi-
ciency in a broad application area of cognitive rehabilitation. Classification accu-
racy was obtained as 92.79% and different other metrics proved that the
combination of channel selection, feature extraction, and classification methods in
our proposed approach has outperformed the others.
Chapter 13: Arduino-based robot for purification of

COVID-19 using far UVC light
The proposed work elaborates on the necessity of sterilization, especially during
the COVID-19 period. In this chapter, we will discuss the development and con-
struction of an “Arduino-based robot for the purification of COVID-19 using far
xxii Smart health technologies for the COVID-19 pandemic
UVC light”. The objective of the current work is to discuss various inventions made
during the pandemic scenario and to contribute to the battle against COVID-19
propagation. Authors have developed a UV-based sterilization method that uses
single-wavelength far-UVC light produced by filtered exclaims to selectively
inactivate microorganisms while causing no biological damage to exposed cells and
tissues. The robot can also be used as a vacuum cleaner, eliminating the need for
and expense of cleaning by humans.
Chapter 14: Effect of COVID-19 pandemic on waste

management system and infection control
Personal protective equipment has become a very vital element to guard against
exposure of any infectious materials. It is usually used in healthcare or hospital
settings during any outbreak. These enormous challenges are with all stakeholders
for avoiding spread through waste. Among all the issues created by COVID-19, the
problem that will create a major mess is, not handling biomedical waste properly. If
the spreading needs to be controlled, it requires strict monitoring of the complete
cycle starting from the point of generation. However, due to the invaluable service
of the waste management sector only, it is possible to avoid the spread of the
COVID-19 and to ensure that the weird tons of waste will not be gathered, which
increases health risks and the spread of disease.
Chapter 15: Natural adjunctive therapies options other than

COVID-19 antiviral therapies
The COVID-19 pandemic is the largest health problem worldwide. Unfortunately,
the lack of an effective and clear treatment causes it to be a major health problem.
There are currently no effective antiviral drugs or vaccines. The symptoms and
course of the disease differ individually. Most of these nutrients also have been
demonstrated to be useful in the treatment of COVID-19. The importance of such
dietary elements in immunity and their particular consequences in COVID-19
patients are discussed in this review.
Chapter 16: Risk assessment and spread of COVID-19

The word risk assessment refers to the procedure of identifying, evaluating the
factors that cause harm to the environment and living beings. It also involves
making decisions on how to put an end to it. This chapter discussed the steps taken
to assess the risk and stop the spread of pandemics like COVID-19. Prevention is
always better than cure. The chapter presented an analysis of prevention using
prediction techniques. Using sentiment analysis in machine learning, weighted
density ensembles, forecasting models, and risk assessment can be done. Further, a
bot named SAUCHA is proposed for automatic sanitization.
Chinmay Chakraborty
Joel J.P.C. Rodrigues
Chapter 1
Internet of Things (IoT) and blockchain-based
solutions to confront COVID-19 pandemic
Abu Hasnat Md Rhydwan1, Md Mashrur Sakib Choyon1,
A.S.M. Mehedi Hasan Sad1, Kazi Ahmed Asif Fuad1,
Kawshik Shikder1, Chowdhury Akram Hossain1 and
M. Shamim Kaiser2
Abstract
COVID-19 pandemic, an unprecedented event that has severely affected every

aspect of human civilization. From the beginning of the pandemic, the contagious
nature of this virus resulted in its rapid transmission throughout the world. As a
result, worldwide health organizations and governments are facing tremendous
pressure to deal with the affected populations. In this demanding period, the
applications of the latest technologies to prevent the spread of the virus are critical.
Among various technologies, the Internet of Things (IoT) and blockchain are being
used in several solutions starting from contact tracing to forecasting. The use of IoT
technologies has proved to be highly effective in such a state of the pandemic.
Conducting real-time health monitoring on patients or suspected cases of COVID-
19, tracking medications, detecting any new suspected cases, diagnosing patients
from a distance, etc. have become exclusively possible with the use of IoT tech-
nologies in this COVID-19 pandemic. On the other hand, the blockchain technol-
ogy that became popular with the increase of different cryptocurrencies has seen its
applications almost everywhere. The technology uses a decentralized system
instead of a single point of contact, proving to be more secure than existing solu-
tions. Blockchain-based solutions are also being used during the ongoing pandemic
in various aspects for secure contact tracing, data handling, and preventing data
fabrication. In this chapter, IoT and blockchain technology are discussed briefly
while describing their core elements. Then, the latest solutions were presented
based on these technologies in different aspects of COVID-19 pandemic preven-
tion. These solutions mainly focus on using these technologies for remote patient
1
Department of Electrical and Electronic Engineering, American International University-Bangladesh,
Dhaka, Bangladesh
2
Institute of Information Technology, Jahangirnagar University, Savar, Dhaka, Bangladesh
2 Smart health technologies for the COVID-19 pandemic
monitoring, secure data handling, and telemedicine. Finally, challenges of using

these technologies were discussed, and possible solutions were recommended to
improve their efficacy in the future.
Keywords: COVID-19; IoT; Blockchain; Contact tracing; Telemedicine;

Smart Healthcare; Smart system; Health monitoring
1.1 Introduction
“Coronavirus Disease 2019” that is mostly known as “COVID-19” is mainly a

highly infectious disease caused by the SARS-CoV-2 virus, otherwise known as the
coronavirus. This virus infects the respiratory system in the human body. The
common symptoms of the disease are cold, fever, fatigue, dry coughs, headache,
loss of appetite, smell, etc. There are also some cases found where the infected
human body showed no symptoms while some others showed severe symptoms like
high fever, shortness of breath, difficulties in breathing, pneumonia, etc. The virus
has spread so rapidly toward the end of 2019 throughout the whole world that the
World Health Organization (WHO) had declared it a “pandemic” [1]. Until the first
week of May 2021, the disease has infected over 153,738,171 people throughout
the whole world while claiming the lives of 3,217,281 people [2]. Modern tech-
nologies throughout the entire world have given a chance to confront the infectious
disease. The rapid spread of the disease required the development and use of
technological resources at a dire speed. The advanced technologies have given
healthcare professionals and patients a great advantage to confront this sudden
pandemic situation.
Among many advanced technologies, the Internet of Things (IoT) and the
blockchain technologies have provided some of the fascinating features that marked
significant contributions in the health sector and fought against COVID-19.
Maintaining physical distance is a crucial aspect of treating a COVID-19 patient; it
has become rather difficult for healthcare professionals to provide service to the
patients in such a situation. However, IoT technologies have enabled health mon-
itoring systems to provide service to the patients remotely, detect possible cases,
deliver logistics, disinfect areas, and carry out telemedicine services for patients [3].
The IoT technologies enable a continuous interaction between devices and physical
surroundings or any subject. Among many operations, healthcare service, public
safety monitoring, and environmental monitoring are some of the commonly used
areas where IoT technologies are leaving huge impacts. Amidst the COVID-19 pan-
demic, the IoT technologies significantly impact the healthcare service by maintaining
medical data, patients’ history, patients monitoring, emergency response, and tele-
medicine service. Because of the rapid and devastating spread of COVID-19 world-
wide, countries and their economic activities have remarkably ceased. However, IoT
technologies can play a crucial part in mitigating a large portion of these issues by
helping carrying out these activities while minimizing the spread of the virus.
IoT and blockchain-based solutions to confront COVID-19 pandemic 3
On the other hand, due to the increasing pressure because of the ongoing pan-
demic, healthcare system of many countries faced catastrophe. The number of
diagnoses of other diseases was significantly reduced [4]. Data management in hos-
pitals and federal repositories has faced unforeseen situations and tremendous risks of
data breaching. As a result, a new challenge has arrived to ensure the privacy of the
database in the healthcare sector. Blockchain technology can provide significant
advantages in this aspect and can maintain privacy in the healthcare sector, especially
when combined with IoT [5]. This technology is being used to increase medical
system security and ensure feasible data sharing [6]. It has also seen its application in
electronic medical records (EMRs), pharmaceutical supply chain, and health insur-
ance claim [7]. In addition to that, blockchain-based solutions are also effective
against other recent issues by maintaining the efficacy of contact tracing, data shar-
ing, Internet of Medical Things (IoMT), and preventing data breaching [8].
In this chapter, different aspects of IoT and blockchain technologies that are being
used throughout the world to confront COVID-19 were discussed briefly. The chapter
gives an overall idea of how these technologies are making significant contributions in
this pandemic by facilitating in different sectors like healthcare, telemedicine, logistics
delivery, surveillance, data management, data privacy, supply chain, securing data-
base, etc. Multiple recent works have also reviewed the application of these technol-
ogies in different aspects of COVID-19 [9–13]. However, these works have focused
on a single technology and excluded some latest applications that simultaneously use
both technologies. Hence, it is necessary to survey applications of both IoT and
blockchain technology, and through this chapter, we tried to cover the existing
research gap. Key contributions of this chapter are as follows:
● We provided an overview of the main components of IoT and blockchain

technology.
● We reviewed the latest IoT and blockchain-based solutions that can be effec-
tive against the COVID-19 and future pandemic scenarios.
● We discussed the significant limitations and challenges of IoT and blockchain
technology and provided deliverables to solve these issues to ensure their
efficient usage.
The rest of this chapter is organized as follows: Section 1.2 contains the
description of these two technologies and their components. Then, Sections 1.3 and
1.4 describe various latest developments of IoT and blockchain-based technology
to reduce the effect of the COVID-19. Then, Section 1.5 addresses various issues of
these technologies, discusses their solutions, and provides recommendations for
their effective implementation. Section 1.6 summarizes the significant findings of
the survey, and finally, Section 1.7 summarizes this study and addresses the future
development of these technologies.
1.2 Internet of Things (IoT) and blockchain overview
This section contains an overview of IoT and blockchain technology.

1.2.1 Internet of Things

The IoT has brought significant advancement in the technological sectors and
engineering throughout the entire world. The IoT is mainly a network of different
embedded entities or objects such as sensors, devices, and databases, and all these
are connected through the Internet. IoT applications basically consist of three
crucial steps. These are data collection, data transmission, and data management.
At the first step, the data is collected by the IoT devices from the targeted envir-
onment or subject utilizing any sensors or devices. Later, these data are transmitted
to other devices, computational units, or data-processing units. Finally, these data
are sorted for thorough analysis, management, extracting meaning or value, making
predictions and decisions. The IoT has enabled access to low-cost technologies,
more effortless connectivity of different entities through the Internet, access
through cloud computing platforms, and push the boundaries with machine learning
and analytics along with artificial intelligence applications [14]. IoT applications
are being explored and deployed in numerous sectors (e.g., enterprises, industries,
healthcare, retails, governments) [15,16]. Therefore, IoT plays a tremendous role in
developing smart cities, smart grids, smart healthcare, smart homes, etc. An esti-
mation suggests that by the year 2025, the number of IoT connections will cross
over 75 billion worldwide [17]. That means the total number of IoT-connected
devices would be more than six times that of the total world population of
humankind. Due to the sudden surge of COVID-19 infections, maintaining physical
distance among humans has become crucial [18]. As a result, the usual flow of
many sectors like healthcare, business, deliveries has become challenging to
maintain. However, IoT applications have given a great chance to mitigate such
issues and offer faster solutions to these sectors. One of the most highly benefited
sectors by IoT is the healthcare sector. The IoMT is also playing an important role
in the healthcare sector in the COVID-19 pandemic [19–21]. Researchers world-
wide have given great effort to work on different projects and applications to
confront the COVID-19 [22–24]. In addition, many tech companies also made
significant contributions in manufacturing some of the advanced technologies to
confront the coronavirus pandemic. As a result, there has been a significant growth
of technological applications in both research sectors and industrial sectors to
confront the COVID-19 pandemic.
1.2.1.1 Major IoT components

This section will provide an overview of the different vital components of IoT.
Figure 1.1 shows the summary of these components.
Things or devices: The most primary requirement for the implementation
of any IoT application is the devices. Every IoT application has its physical
entity, and necessary devices or sensors consist of that entity. The devices
continuously collect the data from the environment or any subject and transfer
those data to the next layer or gateway [14]. IoT applications’ device entities
may consist of various sensors ranging from micro smart sensors to large
industrial machinery.
Analytics
Things
Central
or
control
device
Major IoT
components
User
Security
interface
Cloud Network
Figure 1.1 Components of IoT
User interface: The user interface or UI is the visible entity of an IoT appli-
cation that the users see and provide actions or commands to the application. In
recent times, most IoT applications consist of well-designed UIs that are attractive
and user-friendly. Any IoT-based UI design must have proper hardware constraints
along with the software design. In their work, Toyama and Hirayama presented
some rules and methods in designing such UI for IoT systems [25].
Cloud: The IoT cloud entity is a high-performing sophisticated network of ser-
vers that receives and sends data between the device end and the user end. The cloud
entity is also responsible for preserving data and conducting accurate data analysis.
The cloud also processes the machine learning algorithms and other user commands
or programs. The advanced and highly distributed cloud network architecture of IoT
applications facilitates promising aspects to design, host, manage, and optimize more
futuristic IoT applications with less power consumption and more security [26].
Analytics: In IoT applications, analytics play a crucial role in managing and
improving the entire system architecture. Upon receiving analog data from the
device end, the analytics entity carries out a thorough interpretation, analysis and
provides possible solutions based on the analysis to improve the system. Real-time
intelligent analytics allow system designers or engineers to find any irregularities
and prevent them [27].
Network: Any IoT application must be connected to a good network or
Internet. The device end is controlled by the Internet and maintains the overall
communications between all entities of the system architecture. To develop a smart
city with different IoT applications, it is vital to establish a secure and robust net-
work platform for these applications to rely on [28].
Security: Advancement in the field of IoT is also raising some severe security
concerns regarding the applications of this technology. To develop any efficient
IoT system, it needs to have strong security and should have the feature to maintain
privacy. The initial stage of any IoT application may contain vulnerabilities to fall
prone to cyber threats. However, with standardizations and secured development of
these applications, the IoT world is getting more secured in recent times [29–31].
Central control: The central control entity of IoT systems maintains two-way
data traffic within several different entities or networks. It plays as the central con-
trolling unit in the IoT architecture. With services like the Azure IoT Central, any
developer can establish an efficient infrastructure of the desired IoT application [32].
1.2.2 Blockchain
Blockchain is a distributed ledger that records user data in an encrypted manner [33].
Peers of the blockchain network verify the authenticity of every new block within the
blockchain network. Satoshi Nakamoto, an anonymous person or group of people,
invented this technology in 2008 to create a public transaction system for the cryp-
tocurrency known as bitcoin [34]. Since then, blockchain has become a
trustable solution for digital transactions due to its decentralized structure and
immutability [35]. Apart from cryptocurrency transactions, blockchain has been
widely used in other applications, including energy trading, healthcare, supply chain,
and video games, as shown in Figure 1.2. The continuous growth of blockchain use
cases can be observed from a recent report that shows that the worldwide blockchain
market is likely to reach USD 39.7 billion by 2025 from USD 3.0 billion in 2020 [36].
1.2.2.1 Blockchain components

This section will give a brief idea about different parts of blockchain technology.
Block: Every blockchain network contains numerous blocks chained next to
one another. Each of these blocks includes similar components: block header,
timestamp, Merkle root hash, nonce, previous block hash. These blocks are
immutable, and new blocks require permission from previous blocks within the
blockchain [37]. Thereby, attackers cannot easily add their block to the blockchain
without authorization from others. As shown in Figure 1.3, the blocks are stacked
on top of each other and grow like a chain until the final block is added [37]. The
initial block of this chain is known as the “genesis block.”
Hash algorithm: A function that transforms an input of any length into an
encrypted output of a fixed length using a mathematical function is known as the
hash function [38]. This ensures that security as decrypting fixed-length output
without knowing the input length is a complex and time-consuming operation.
Header: A block header is used to identify a particular block on an entire
blockchain. Every block has a unique header, and they are used for documentation
and efficient task recording. They contain three separate sets of block metadata,
including version number, previous block hash, Merkle root, timestamp, and dif-
ficulty target. Information like previous block hash and Merkle root are used to
navigate specific bytes within the blockchain easily.
Cybersecurity Cloud storage

Protection against Secured storage facility
cyberattacks through through decentralized
data immutability structure
Supply chain IoT

Faster tracking of Protection of IoT
products by devices from
retailers and cyberattacks like
customers denial-of-service
attack (DDoS)
Energy trading Transaction

Trusted medium Transaction facility
for peer-to-peer without paying
energy transaction commission to
third parties
Banking Healthcare
Faster and Reduction of risk
secure of data fabrication
payment method and better transpar-
ency of data
between patients
and doctors
Figure 1.2 Current applications of blockchain technology
Block header Block header Block header

Merkle root hash Merkle root hash Merkle root hash
Timestamp Timestamp Timestamp
Nonce Nonce Nonce
Version Version Version
Target difficulty Target difficulty Target difficulty
Previous block hash Previous block hash Previous block hash
Block 1 Block 2 Block 3
Figure 1.3 Blockchain structure
Timestamp: Timestamp is used to prove the approval time of a block. It also

helps block authentication.
Version: The version number helps to track the changes throughout the pro-
tocol. There are various versions based on their applications. Version 1.0 is used for
cryptocurrency, version 2.0 for smart contracts, version 3.0 for DAPPS, and version
4.0 for the industry.
Merkle root: Merkle root checks corruption and manipulation within the
block using mathematical formulas. The unique hash value is created by following
the concept of binary tree and combining the transaction data.
Difficulty target: Every blockchain network has a fixed difficulty level that
miners need to achieve through their computational power. The difficulty target
changes based on the number of blocks that have been added to a network in a
specific amount of time.
Nonce: Nonce, the abbreviation of “Number only used once,” is a 32-bit
number that contains in every encrypted block. Blockchain miners try to solve this
number to receive a reward in the form of cryptocurrency. It has a total of 232
possible values, which makes it computationally expensive for the miners.
Previous hash: In a blockchain network, every block relates to the previous
block of the chain. Every block stores the previous block’s hash value except the
genesis block, which contains no previous hash value.
1.2.2.2 Blockchain consensus mechanism

Every blockchain network follows some predetermined fault-tolerant protocols, to
which every participant of the network must agree upon. This protocol is known as
a consensus mechanism. Blockchain is a decentralized technology; there is no
specific authority to set necessary acts for the network. Using a consensus
mechanism, the networks ensure that every transaction takes place from a legit-
imate source with the consent of other network participants [39]. Various consensus
mechanisms have been emerged in recent times based on the application of
blockchain networks. Among these, proof of work (PoW), proof of stake (PoS),
delegated proof of stake (DPoS), practical byzantine fault tolerance (PBFT), and
ripple are being used the most. A comparison between these mechanisms is shown
in Table 1.1. As the table shows, these consensus mechanisms differ from one
another based on various network requirements, including node management,
transaction cost, and power consumption [38,40–42].
1.2.2.3 Types of blockchain

Since its advent in 2008, blockchain has changed a lot due to its application in
different industries. As a result, the decentralized structure of the earlier blockchain
network has also changed into more privacy-preserving and secure ones [43].
Table 1.2 shows the difference between three types of primarily used blockchain
[43–46]. Private and consortium blockchain seems more viable for business pur-
poses due to factors like transaction speed, privacy, and ownership. However, these
two blockchain deviates from the fundamental quality that makes blockchain spe-
cial—decentralization. But they also eliminate the scalability, power, and transac-
tion issues of the public blockchain network. A hybrid between public and private
blockchain, which is known as a consortium or federated blockchain, can offer the
perfect spot by offering the benefits of both of them [46].
Table 1.1 Comparison between various blockchain consensus mechanisms
Property PoW PoS DPoS PBFT Ripple

Node manage- Open Both Open Permissioned Open
ment
Transaction rate Low High Medium High High
Power con- Large Less Less Negligible Negligible
sumption
Scalability Good Good Good Bad Good
Storage con- High High Medium High Medium
sumption
Transaction High High Medium Low Low
cost
Adversary tol- 25% of com- <51% of <51% faulty <33% Vot- <20% of de-
erated power putational stake replicas ing power fective
power nodes
Application Public Public Public Permissioned Permissioned
Table 1.2 Difference between various types of blockchain
Property Blockchain type
Public Private Consortium

Network Decentralized Partially decentralized Hybrid
Privacy Open Permissioned Permissioned
Ownership Open Single entity Multiple entities
Consensus mechanism PoS/PoW Voting Voting
Transaction speed Slow Fast Fast
Decentralization Completely de- Less decentralized Less decentralized
centralized
1.3 IoT technologies to confront COVID-19
In recent times, applications developed with IoT and IoMT technologies have
become significantly popular among many sectors, and security measures like
the blockchain, edge computing, and fog computing, available for these appli-
cations make them more efficient [39]. The rapid developments made in the
IoMT have made healthcare professionals realize the sector’s vast potential
[47]. Figure 1.4 shows the summary of IoT-based solutions in different sectors
against the COVID-19 pandemic. As shown in Figure 1.4, IoT technologies
have given great advantages in several sectors to confront the COVID-19
pandemic.
Detection of
Telemedicine
IoT in healthcare possible cases or Disinfecting area Logistics delivery
service
patients
Secured data
Surveillance preserving
Real-time health
camera to detect IoT-based supply
monitoring
physical Drones with chain redefined Wider coverage of
distancing disinfecting model healthcare service
Emergency spray
response Smart Maintaining
thermometer to Contactless physical distance
detect delivery with
temperatures of drones
Smart predictions Provide healthcare
human
at the remote area
Drones for Robots with UV
Smart guidance human technologies Autonomous Remote monitoring
and temperature and delivery vehicles through sensors and
recommendations crowd detection
devices
Figure 1.4 IoT solutions in various aspects of COVID-19
1.3.1 Health monitoring systems

IoT-based technologies and other technologies related to IoT have proven to be
significantly helpful and efficient in the battle against the COVID-19 pandemic.
These technologies can provide real-time health monitoring data, physical exer-
cise checks, blood pressure, disease condition, and many other data of the human
body that prove to be extremely valuable in fighting the coronavirus [48]. The use
of IoT has increased the smartness of healthcare by efficiently collecting data
from numerous sources. This has resulted in prompt and more accurate health
monitoring and can also be beneficial for COVID-19 monitoring [49]. An oper-
ating system for information kiosks was developed in [50] that can monitor users’
health in real time by using different sensors and presenting those data on the
kiosk screen, which can be beneficial for crowded locations. Choyon et al.
developed an IoT-based health monitoring system that combines real-time mon-
itoring facilities through hardware and COVID-19 prediction programs through a
machine learning algorithm [51]. The researchers developed the hardware device
with a Raspberry Pi microcontroller, Arduino Uno, and other necessary sensors to
monitor patient’s health. On the other hand, the machine learning algorithm
carries out its analysis on a cloud-based server where the data collected from the
device is sent and stored. The system can predict the possibility of coronavirus
present in a patient’s body by analyzing the collected data from the symptoms.
The developed device can also deliver guidance from healthcare professionals to
monitor the state of a patient’s health from a distance. In addition, in an emer-
gency, the response team can quickly take actions based on the data provided by
the device and mitigate the damage.
Other wearable devices like Fitbit or Apple Watch used for regular health
fitness tracking have also shown some promising features that could be used to
sense early signs of COVID-19 symptoms [52]. Similarly, another successful
gadget that made an essential contribution against COVID-19 is the WHOOP
strap. A study was conducted with the WHOOP strap on 271 people between the
age of 25–50 years. All these people experienced symptoms like that of SARS-
CoV-2, but 81 individuals were tested positive for the virus in later days. The study
consisted of building a dataset collected by the WHOOP strap and training a novel
algorithm using those data. Later, the developed model monitored the raspatory
rate of the individuals and successfully detected 20% of the people 2 days before
the symptom commencement and 80% of the people on the third day [53].
In another work in [54], an IoMT system was developed that utilizes deep
learning for identifying different types of health-related issues caused by COVID-
19. The system generates reports based on the analysis that can be used for medical
care. Bai et al. researched the IoT-based nCapp technology to diagnose COVID-19
at an earlier stage [55]. Based on the collected data, the diagnosis automatically
labels the patient’s case as confirmed, suspected, or suspicious of the coronavirus
infection. The nCapp intelligent assisted management can play a crucial role in
treating the infected patients through various instructions based on data analysis
collected from the patient’s body. A health monitoring model was developed based
on the Business Process Management paradigm, where the main focus was to
increase the empowerment of the patients and healthcare professionals [56]. The
architecture of the model was developed with the help of health monitoring devices
and environmental sensors. Together with the connection of the web system, an
efficient IoT technology was developed for the model that could fulfill the purpose.
The patients could monitor their state of health and contribute to their health rou-
tine by deciding what to do and when to do it. Lomotey et al. developed a “Petri
Nets” service model that focuses on wearable IoT-based device data management
[57]. The proposed model by the authors has resulted in high scalability of the IoT
architecture, transparency of medical IoT data, and distributed health information
system threats. In [58], an IoT-based smart system was developed for monitoring
the health of patients. The patients can use the system through a mobile application,
and it gives the patients the freedom to monitor their health from home by them-
selves and also enables them to send necessary health reports to the doctors through
the system. Furthermore, the doctors can provide necessary health guidance to the
patients through the system as well. The system was proven to be quite helpful for
elderly people or people living independently at home. In [59], an IoT-based health
monitoring system was developed utilizing three core parts: IoT node, mobile app,
and machine learning tools. The developed system also utilizes a radio frequency
method to alert the user to maintain physical distance in both indoor and outdoor
environments. In another work, Li et al. designed a health monitoring system that
could analyze the breathing sound in real-time and detect any wheeze in the
breathing [60]. As the lungs are one of the most commonly affected parts of the
human body due to the coronavirus, such a system could prove highly useful for
healthcare professionals in detecting any blockage created in the lungs by the virus.
In another work, de Morais Barroca Filho et al. proposed an efficient IoT-based

healthcare platform where patients in critical conditions can be monitored remotely
[61]. The researchers, in this case, utilized wearable and unobtrusive sensors to
monitor patients in ICUs.
1.3.2 Tracking and detecting possible patients

IoT-based smart thermometer developed by a health technology company called
Kinsa from the United States has a significant advantage in detecting people with
high body temperature [62]. The service can be used on a massive scale and detect
potential hotspots of COVID-19 affected areas. In terms of COVID-19 surveil-
lance, various advanced methods are now being used in different parts of the world.
In [63], a novel IoT-based occupancy system was developed that enables the
facility to detect the level of occupancy in any area such as buildings, public space,
classroom, office room, and transportations. The developed system is also secured
with the blockchain while containing a decentralized traceability. The developed
system does not record any unauthorized personal data, and therefore it is also
proven to be secured while maintaining privacy. In China, drones called
“MicroMultiCopter,” manufactured by the Shenzhen company, carry out surveil-
lance to detect any crowds, detect body temperature, ensure mask-wearing by the
civilians, etc. In addition, drones are also being used for contactless products or
medicine delivery in various places [64]. In Madrid of Spain, drones are also being
used for delivering important COVID-19-related guidelines or emergency
announcements to the citizens from the authority [65]. In Australia, the University
of South Australia developed an advanced drone associated with a Canadian
company called Dragonfly Inc. The drone could carry out surveillance on a large
area while detecting humans with infectious diseases. The drone could detect the
temperature of the human body and use image processing to detect heart rate [65].
Since maintaining the physical distance from one person to another is vital in
fighting against COVID-19, the IoT technologies with the combination of computer
vision (OpenCV) and deep learning can be of great help to law enforcement
agencies in maintaining it [66]. The surveillance camera placed around the corner
of streets can monitor the movement of the people and thus detect whether people
are maintaining or violating physical distance among them during the COVID-19
pandemic. As per the developed algorithm, the system detects those who maintain a
safe physical distance from others and mark them with an identifying-colored box
on the footage (e.g., green box). On the other hand, those who do not maintain a
safe physical distance are marked with another colored box on the footage (e.g., red
box). The local law enforcement agency gets notified of the activity as well to take
specific measures if required.
In another work, Abdulrazaq et al. developed an IoT-based novel smart hel-
met that could detect possible COVID-19-affected people from the thermal image
taken by the camera mounted with the helmet [67]. The device has a great promise
to be fruitful in detecting suspected cases in the public or hospitals. The device
detects any suspected case with higher body temperature, and it sends a
notification to the paired mobile, any other device, or database with the captured
face image and GPS location. Based on the received information, the healthcare
professionals or any other concerned party can take necessary actions. Similarly,
IoT-based smart glasses were developed to fulfill the same purpose [68]. In [69],
the researchers aimed to design an IoT-based global platform to confront COVID-
19 through monitoring and tracking the suspected cases. The system utilizes a
virtual IoT node to determine the suspected cases. The system was also designed
to ensure the privacy and authenticity of information providers. The information
providers are given with individual secret keys to ensure the authenticity and
privacy of data. Garg et al. reviewed different contact tracing models, and later on,
they presented a novel contact tracing model that proved to be scalable and can
also maintain the patient’s privacy [70]. The developed model utilizes IoT and
blockchain technologies that can maintain a greater efficiency in contact tracing of
COVID-19 cases while also considering moving objects with the help of RFID
transceivers. In [71], a privacy-preserving IoT model was presented that can
efficiently identify a large number of infection contacts while preserving the
individual’s privacy with the help of blockchain. In another work, Tedeschi et al.
developed IoT-based contact tracing architecture that could preserve privacy while
being configured to support several models ranging from centralized to decen-
tralized [72]. Furthermore, to impose strict lockdown on any COVID-19 hotspot
area, an IoT-based three-layered decentralized biometric face detection framework
was designed in [73], where the authors built a deep learning framework. The
framework showed significantly better results when compared with other existing
state-of-the-art architecture.
1.3.3 Disinfecting area

IoT-based robots are also proving to be quite helpful during the COVID-19 pan-
demic. Robots are deployed in many advanced countries to disinfect areas, detect
affected patients, carry out contactless delivery, and other diagnosis or screening
purposes. The new generation robots of macro- to microscale also enable more
significant advantages to navigate or sterilize high-risk areas [74]. A USA-based
company called Xenex manufactures Germ-Zapping robots that use UV lights to
destroy any kinds of germs quite efficiently [75]. The company has reported
attaining colossal success and demand in various countries worldwide like Italy,
Singapore, Japan, and South Korea to fight against coronavirus. Another robotic
company from Denmark called the UVD Robots had developed a highly effective
remote-controlled robot that uses powerful UV light to disinfect the surface. The
UV light focuses on destroying the DNA of the viruses or bacteria directly. Each
robot advances 3 m every minute while disinfecting the area with the UV. The
company has already produced robots in many hospitals in different parts of the
world like Asia, Europe, and the United States [76]. Furthermore, drones enable
quite an advantage in disinfecting any public area as well. China, Indonesia, Chile,
Spain, and many other European countries use drones to disinfect public areas
amidst coronavirus pandemic [77].
1.3.4 Telemedicine
The telemedicine service has seen a significant surge in usage throughout the world,
especially in this COVID-19 pandemic. The demand for medical counseling has
been higher than before while maintaining social distance was an essential condition
to abide by. In such a circumstance, telemedicine service proves to be a great help in
delivering medical services to the patients to a certain extent. The doctors or
healthcare professionals provide their expert guidance to the patients through tele-
medicine service through telecommunications equipment with the minimum facility
of audio and video [78]. In some cases, necessary sensors or biomedical devices that
can collect patients’ biological data are also installed [79]. As a result, the demand
for telemedicine services has gone up significantly in the COVID-19 pandemic [80].
The healthcare professionals could examine the patients’ data from their mobiles or
desktops and provide necessary instructions to the patients or medical staff taking
care of the patient [81]. With the help of telemedicine service, the massive pressure
of providing medical service by the hospital staff got lifted, along with the risk of
coronavirus emanation was reduced. In the United States, the George Washington
University hospital and Rush University Medical Center utilized telemedicine ser-
vices to provide medical consultation to their patients during the pandemic [82,83].
These hospitals arranged one-to-one video calls and webinars to provide medical
consultations. The German-made WAS vehicles are equipped with advanced IoT
technologies that can enable medical staff to monitor a patient’s health remotely
[84]. Therefore, the emergency vehicle can provide healthcare support even on the
move to the hospital. The 5G technologies combined with IoT and other related
technologies can also influence carrying out swift telemedicine service [85,86].
Such a highly advanced technology could perform critical actions like surgeries on
patients through robotic devices while being controlled by the doctors remotely.
1.3.5 Logistics delivery

The use of autonomous vehicles has also been quite promising in logistics delivery
without the risk of human exposure to the coronavirus [87]. These autonomous
vehicles have been quite successful after being deployed in several parts of the
United States and China. Končar et al. analyzed the setbacks of digitalized business
processes and the fast-moving consumer goods supply chain based on IoT tech-
nologies [88]. The authors carried out the research during the COVID-19 pandemic
and identified some of the lackings that usually occur due to human errors or
capabilities. The authors proposed that the traditional business models and supply
chain should be changed with the help of intelligent systems and IoT technologies
to regain stability, especially during a pandemic like COVID-19.
1.4 Blockchain technologies to confront COVID-19
Blockchain-based technologies are being used everywhere to improve the current

technologies of pandemic handling. Most of these applications are related to
security improvement, which is a massive issue in the current scenario. Figure 1.5
Contact
tracing
Internet of
medical
Things
(IoMT)
Database
Blockchain security
solutions
for
COVID-19
Prevention
of data
fabrication Data
sharing
Figure 1.5 Use of blockchain technologies in various perspectives of COVID-19
shows different perspectives of blockchain-based solutions against COVID-19.

This section will highlight the recent developments in these sectors.
1.4.1 Contact tracing

From the initial stage of the COVID-19 pandemic, the governments of different
countries depended on contact tracing to reduce the spread of the virus. However,
some obstacles are reducing the efficiency of this method: (1) patient information
stored in centralized databases is prone to tampering. (2) Existing infection tracing
systems consisting of location-based contact tracing and individual contact tracing
cannot track infections from different dimensions [89]. While location-based
tracking provides information within a given area without information on the
movement of the infection, individual tracing focuses on person-to-person tracking
without any data on the area where the infection takes place. To solve these issues,
third-party servers are required to check contact and sending alerts. Furthermore, in
[90], a contact information and risk notification sharing system was designed using
smart contracts and Bluetooth that can eradicate this issue by recording the visited
location of the users and providing the option to update their infection status. The
system can also update infection status and estimate the probability of being
infected based on the user’s visited places. In another contract tracing solution, a
public blockchain network is used as a distributed public ledger to allow the user to
securely register, upload and query their contact list in the blockchain network [69].
In another work, a blockchain-based system was developed for authorities to pro-

mote social distancing in society by allowing specific individuals to visit any
location in an area [70]. In this work, blockchain was used for conducting privi-
leged information transfer among trusted participants.
Blockchain increases the trustworthiness of contact tracing applications by
keeping every block visible to the network participants. As a result, it has turned
into a trusted medium for digital contract tracings (DCT) applications such as
BeepTrace and BeepTrace-Active [91,92]. In these applications, blockchain was
used as the primary infrastructure for DCT. In [91], the use of blockchain was
proposed to work as a bridge between user and authority to secure the user’s ID and
location information. While this application faces computational power and com-
plexity issues and uses passive positioning methods, Klaine et al. [92] used geo-
graphical data and Bluetooth to track contact tracing. This work also allows
tamper-free access to perform local matching, which increases user privacy.
Finally, a blockchain-based framework was proposed in [93] to investigate the use
of various blockchain-based technologies like peer-to-peer transactions, time-
stamps, and decentralized storage to increase the effectiveness of COVID-19
detection.
1.4.2 Database security

One of the core strengths of blockchain networks is their ability to decentralize
information, which removes existing technologies like database and cloud, which
has a single point of failure [94]. In [95], the authors proposed blockchain tech-
nology for data storage and management in hospitals to securely store patient data
and track affected patients. Blockchain ensures the digitalization of data and
enables more accessible access to everyone, which eventually leads to the detection
of more potential cases. To ensure that, Manoj et al. [96] developed a unique
incentive-based approach to prevent information tampering within a database.
Many recent works focused on developing a secured database for storing patient
information. Choudhury et al. proposed the use of blockchain for secure record-
keeping of different personal information like age and medical records to ensure
anonymity [97]. The data within the records would be available on a read-only
basis to the stakeholders (state health centers, government, hospitals). In another
work, a smart contract-based Ethereum blockchain was developed to prepare
digital medical passports for COVID-19 test takers. The work has reduced the
response time of healthcare facilities for the users and reduced the extent of fab-
ricated information [98]. Similarly, Resiere et al. implemented medical cooperation
in the Caribbean using blockchain technology to protect patient data, ensure
facilities like patient referral, and establish new payment standards [99].
1.4.3 Information sharing

COVID-19 pandemic has also significantly raised the necessity of gathering the
latest and reliable data related to the virus spread. Blockchain has the ability to do
that as this technology allows to verify data easily using the distributed ledger and
peer-to-peer networking facility [70,100]. Many recent platforms are also using this
technology to record patient information with high-security symptoms, location,
and medical records. The WHO has also launched a blockchain-based platform
called MiPasa that allows private information sharing between its users, health
organizations, and authorities. The platform also has a self-reporting feature using
which the users can report their infection with public health organizations with
precise locations [101]. In another work, a blockchain-based framework was
developed for exchanging health information [102]. The framework also allows the
patients to control their medical information and to preserve personal autonomy.
Also, in [103], a decentralized healthcare system was developed using blockchain
technology to encourage patients to safely share their medical information with
health organizations while retaining complete control of their data privacy. The
framework has the potential to solve many existing challenges of the conventional
healthcare system by offering data protection, less operating cost, and easier access
to data. A blockchain-based method was built to ameliorate data transmission and
improve the reliability of physiological information [104]. The method protects
recorded data from node manipulation threats and enables secure information
sharing between multiple parties.
1.4.4 Prevention of data fabrication

Access to uncorrupted datasets is one of the critical requirements while researching
various diseases, including COVID-19. However, storing the data in a conventional
database can result in tampering. In addition, cases of medical data fraud are also
rising in recent days [105]. For instance, the result of a recent report on the frontline
workers of Pakistan shows that data falsification by the vaccinators is a typical case
[106]. In this regard, blockchain can help with its decentralized structure and other
features like a smart contract. Several recent literature studies have focused on the
necessity of good data for prediction on the pandemic spread and prevention and
raised the question of the quality of existing database management systems
[95,107]. The tracking system presented in [107] addressed the data manipulation
issue and removed the single-point failure issue. In another work, an incentive-
based technology was introduced using blockchain technology to prevent tamper-
ing in COVID-19 test results [96]. Similarly, in [108], the possibility of causing
panic in public during the pandemic due to tampered data was discussed. The
authors also investigate the possibility of poor scientific results due to working with
falsified data. The work of Ramachandran and Kantarcioglu showcases a
blockchain-based data provenance system that can store encrypted data and can
trigger smart contracts in the case of any sort of data fabrication [109]. A
blockchain-based framework was proposed in [110] to increase the security of
EMRs. The framework also uses a credit score system to ensure the credibility of
authorities who enters data into the blockchain. Finally, in [111], a Merkle-tree-
based approach was followed to protect the integrity of EMR while removing
complex features of blockchain like mining. This method can be extremely useful
of pandemic data handling due to the simple approach followed in that work.
1.4.5 Internet of Medical Things

There have been numerous recent applications of the IoMT in healthcare due to the
advances in sensors and medical devices. This technology also has massive potential
in pandemic handling. However, IoMT has also faced some privacy concerns due to
the vulnerabilities of the biomedical sensors. These sensors are often resource-limited
and prone to wiretapping, malware, and word attacks [112]. Blockchain can improve
this issue by improving authentication, and access control on various layers of
blockchain enables IoMT. Dai et al. presented a blockchain-enabled IoMT solution
and discussed the privacy concerns of IoMT devices [113]. In another work, the
combination of IoMT application and blockchain was illustrated for efficient patient
data analysis [114]. A remote patient monitoring system was proposed in [115] that
uses blockchain to overcome the security concerns of IoT devices. In [116], the
authors proposed a blockchain system using the Ethereum platform where IoMT
devices can be used in carrying out telemedical laboratory examinations. The system
was proven to be efficient in both time and cost. Another work in [117] introduced
IoT device monitoring architecture with the help of a private blockchain. This
architecture can detect changes to the IoT device architecture and can ensure a fea-
sible and secure management of IoT device in large networks.
Our results show that such a system is possible, and the blockchain can secure
the dissemination of configuration changes to IoT devices. The key novelty of our
solution is a distributed management of configuration files of IoT devices in
enterprise networks utilizing blockchain technology.
Telemedicine, a technology that enables features like remote patient monitor-
ing, diagnosis, and treatment at a lower cost, is highly effective in the current
situation [118]. As COVID-19 is a highly contagious virus, it is highly challenging
for frontline workers to aid patients while keeping themselves safe. Various IoMT
devices like wearable devices and advanced sensors are a crucial part of tele-
medicine throughout the years. However, this technology is also prone to cyber-
attacks and data breaching [119]. There are many recent works where blockchain
was used to make the IoMT devices more secure for their effective use with tele-
medicine. In [120], a blockchain-based framework was proposed that uses block-
chain technology to improve the security and efficiency of telemedicine services. In
another work in [121], an attribute-based encryption scheme was presented that can
ensure higher efficiency for on-demand medical services in telemedicine system.
1.5 Challenges, solutions, and deliverables
In this section, different challenges that obstruct the development of IoT and
blockchain applications are mentioned. Furthermore, some practical solutions and
recommendations to mitigate these issues are also briefly discussed.
1.5.1 Challenges of IoT and blockchain technology

With any technology, there are always some issues and challenges that need to be
mitigated. In the case of IoT applications, the devices are prone and vulnerable to
cyberattacks. Cyberattacks on IoMT applications can cause severe damage to

patients’ lives as some of these applications mount devices directly to the patient’s
body. Furthermore, IoT applications are closely related to some sensitive data of
the subjects. In terms of IoMT, there is a considerable amount of data consisting of
patients’ information. Therefore, the sensitive data and privacy of people using
these IoT-based applications are vulnerable to cyberattacks. Some of these attacks
may even result in the identity theft of patients. IoT applications are vulnerable to
various types of cyberattacks such as botnets attack [122], DoS attack [123], DDoS
attack [124], and Eavesdropping attack [125]. Clark et al. showed how botnets
attack could even lead to physical damage to human patients [126]. The authors
also elaborated how IoMT devices may cause psychological damage to patients due
to being surrounded by machines and devices instead of human beings. There are
also some concerns regarding IoT technologies by the user communities. People
are concerned by the possibility that robots might take over their jobs in the coming
days, resulting in trust issues on these applications [127]. There are also concerns
regarding the accuracy of sophisticated IoMT applications. Furthermore, the lack of
standardization of IoT technologies has put a massive concern among users,
developers, researchers, and engineers [128]. Therefore, to implement and confront
a pandemic-like COVID-19 with IoT-based solutions more efficiently, these issues
must be addressed and resolved [20].
In the case of blockchain, there are still some issues and challenges that need to
be addressed before implementing the technology against COVID-19. Due to the
increasing number of transactions, the traffic data becomes bulky, and it affects the
creation of a new block due to the time interval set by the blockchain networks.
Another issue is the processing speed of primarily used blockchain platforms as they
process very few transactions per second and not suitable for processing millions of
real-time transactions. High computational power usage is another huge caveat for
implementing blockchain-based platforms. In the case of PoW, to add a block in the
blockchain, a user has to solve a problem that requires enormous computational
power. Therefore, it would not be suitable for resource-limited cities. Moreover,
blockchain is still pretty complex, and it will require time before people get used to
the technology [129]. Lastly, even though blockchain removes most of the security
problems of conventional databases and cloud, it is prone to various cyberattacks
such as Sybil attack, eclipse attack, 51% attack, DDoS attack, long-range attack, and
P þ epsilon attack [37]. The attacker can take over the entire blockchain network
through these attacks and significantly slow down the operation.
1.5.2 Possible solutions and deliverables
Worldwide, researchers are working to mitigate the cybersecurity issues of IoT
devices. In [130], the researchers proposed an Anomaly detection system that was
developed using a random Forest machine-learning algorithm. The proposed sys-
tem has the ability to find compromised IoT devices in a distributed Fog Nodes. In
another work, researchers utilized IoT HoneyNets in order to generate distractions
for the attackers from their desired targets and to mitigate cyberattacks [131].
Recent works also focused on the psychological attacks suffered by humans due to
leakage of their personal data through IoT devices, and to mitigate those issues, and
various frameworks were developed [132,133]. On the contrary, recent attacks on
blockchain have raised the question of the immutable nature of the blockchain
network. To prevent future cyberattacks, numerous recent works have focused on
mitigating those attacks, including mitigating pool hopping attacks, false data
injection attacks, and 51% attacks [37,134,135]. A parallel PoW network was
proposed in [104] instead of solo mining to increase the processing speed of
transactions in the blockchain network. The proposed method can also improve the
scalability issue of blockchain. In another work, various versions of blockchain
were reviewed to find the optimal versions that require less computational power
[136].
Without resolving the crucial issues of IoT applications, the public acceptance
of these services will not increase significantly. First, the data of these applications
must be adequately encrypted. Without proper encryptions, the systems might fall
under cyberattacks, and they might also get beyond recovery. Also, the privacy of
personal data of any patient, subject, or human must be ensured in these applica-
tions. Since some of the attacks might occur through phishing, the people asso-
ciated with the system should be trained appropriately and raise their technical
knowledge or awareness. Also, the staff in charge of maintaining these applications
should be specialized to have sufficient knowledge in maintaining, safeguarding,
securing, and developing the systems. Increasing the trust and awareness of users or
patients about IoT applications is another aspect that needs to be ensured. In the
end, standardizations of these IoT applications should be ensured to have efficient
implementations of these applications.
To effectively use the blockchain-based solutions against COVID-19, specific
steps are required to be taken. Network latency of the blockchain network needs to
be improved to handle extensive data, and a suitable blockchain platform is
required to use based on the application to handle large transactions. Besides,
security issues need to be prioritized to increase user trust in such applications. To
do that, available latest technologies should be implemented against various
cyberattacks. Finally, energy consumption issues need to be handled to ensure the
large-scale use of blockchain-based platforms.
1.6 Key findings and discussion

In this section, the key findings from the conducted survey have been mentioned,
and these applications’ strengths and limitations have been discussed.
As shown in Table 1.3, IoT-based applications have made a huge impact on
ensuring contactless and remote healthcare monitoring. During the ongoing pan-
demic or such future events, these applications can reduce the damage sig-
nificantly. In addition to that, these applications are primarily user-friendly and
cost-efficient. Therefore, it can be used in almost every area. However, these
applications are prone to different security threats that can mitigate their efficient
usage. On the other hand, blockchain has ensured a secure and trusted environment
Table 1.3 Key findings from reviewed IoT and blockchain-based solutions
Technology Applications Strengths Limitations References

Internet of ● Health monitor- ● Enables health- ● Requires specific [41–81]
Things ing systems care service knowledge and
(IoT) ● Tracking and remotely without training to
detecting possi- any physical maintain
ble patients contacts ● Less security
● Disinfecting area ● Broader coverage could lead to
● Telemedicine of service becoming vulner-
● Quicker response able to cyber-
● Logistics threats
delivery in emergencies
● Real-time ● Maintenance and
monitoring checking required
● Safe and clean
environment
● Always available
rooms for
improving or add-
ing features
● User-friendly
● Low-cost projects
Blockchain ● Contact tracing ● Increased security ● Lack of user [82–114]
● Database over traditional awareness of the
security databases technology
● Information ● Trusted platform ● Slower perfor-
sharing for COVID-19 mance in com-
● Prevention of application parison to cloud-
data fabrication development based services
● IoMT ● Rise of trustability ● High computa-
among users tional
● Improvement of power usage
remove healthcare ● Not suitable for
monitoring resource-
limited cities
for application development despite being a comparatively new technology.

Blockchain-based solutions are secure and can ensure trustability during among
users. Nevertheless, these applications use substantial computational power and can
be inappropriate for small cities.
1.7 Conclusion and future scopes

In this study, different applications of IoT and blockchain-based technologies that
can be effective against the COVID-19 pandemic were presented. In this demanding
time, these solutions can help to slow down the spread of the virus. The study
includes a brief investigation of different core components of these technologies to

understand their potential in the health sector. It was found that while IoT-based
solutions can solve issues like patient detection, health monitoring, and telemedicine,
blockchain-based applications can be used to ensure a privacy-preserving environ-
ment. However, these technologies have some caveats requiring consideration and
should be solved before their widespread use for pandemic handling. The future
scopes for these technologies are enormous as applications of both technologies are
expanding rapidly. Systematic development of these applications can significantly
improve their efficiency and trustability in the future. In addition, with efficient
implementations and resolving some of the drawbacks of these technologies, con-
fronting any pandemic like COVID-19 or a similar difficult situation can be more
effective.
References
[1] Harvard Health Publishing. COVID-19 Basics [online]. 2021. Available

from: https://www.health.harvard.edu/diseases-and- conditions/covid-19-
basics [Accessed May 2021].
[2] WHO. Coronavirus disease (COVID-19) pandemic [online]. 2021. Available
from: https://www.who.int/emergencies/diseases/novel-coronavirus-2019
[Accessed May 2021].
[3] Pittaway, D. The Use of IoT During COVID-19 [online]. 2020. Available from:
https://www.iotforall.com/use-of-iotduring-covid-19 [Accessed May 2021].
[4] Dinmohamed, A.G., Visser, O., Verhoeven, R.H., et al.: ‘Fewer Cancer
Diagnoses During the COVID-19 Epidemic in The Netherlands’ The Lancet
Oncology, 2020, 21(6), pp. 750–751.
[5] Liang, X., Zhao, J., Shetty, S. and Li, D.: ‘Towards Data Assurance and
Resilience in IoT Using Blockchain’, in ‘MILCOM 2017–2017 IEEE
Military Communications Conference (MILCOM)’, (2017), pp. 261–266.
[6] McGhin, T., Choo, K.-K.R., Liu, C.Z. and He, D.: ‘Blockchain in Healthcare
Applications: Research Challenges and Opportunities’ Journal of Network
and Computer Applications, 2019, 135, pp. 62–75.
[7] Agbo, C.C., Mahmoud, Q.H. and Eklund, J.M.: ‘Blockchain Technology in
Healthcare: A Systematic Review’ Healthcare, 2019, 7(2), p. 56.
[8] Kaiser, M.S., Mahmud, M., Noor, M.B.T., et al.: ‘iWorksafe: Towards
Healthy Workplaces During COVID-19 with an Intelligent Phealth App for
Industrial Settings’ IEEE Access, 2021, 9, pp. 13814–13828.
[9] Nasajpour, M., Pouriyeh, S., Parizi, R.M., Dorodchi, M., Valero, M. and
Arabnia, H.R.: ‘Internet of Things for Current COVID-19 and Future
Pandemics: An Exploratory Study’ Journal of Healthcare Informatics
Research, 2020, 4(4), pp. 325–364.
[10] Javaid, M. and Khan, I.H.: ‘Internet of Things (IoT) Enabled Healthcare
Helps to Take the Challenges of COVID-19 Pandemic’ Journal of Oral
Biology and Craniofacial Research, 2021, 11(2), pp. 209–214.
[11] Ndiaye, M., Oyewobi, S.S., Abu-Mahfouz, A.M., Hancke, G.P., Kurien, A.M.
and Djouani, K.: ‘IoT in the Wake of COVID-19: A Survey on Contributions,
Challenges and Evolution’ IEEE Access, 2020, 8, pp. 186821–186839.
[12] Nguyen, D.C., Ding M., Pathirana P.N., and Seneviratne A. ‘Blockchain and
AI-based solutions to combat coronavirus (COVID-19)-like epidemics: A
survey’ IEEE Access. 2021, 9, pp. 95730–95753.
[13] Sharma, A., Bahl, S., Bagha, A.K., Javaid, M., Shukla, D.K. and Haleem, A.:
‘Blockchain Technology and Its Applications to Combat COVID-19
Pandemic’ Research on Biomedical Engineering, 2020, pp. 1–8.
[14] Oracle. What is the Internet of Things (IoT)? [online]. 2021. Available from:
https://www.oracle.com/internet-ofthings/what-is-iot/ [Accessed May 2021].
[15] Reshetchenko, D. Top Five Sectors of IoT: Use Cases and Security [online].
2018. Available from: https://dzone.com/articles/5-sectors-which-use-the-
iot-efficiently- but-what-a [Accessed May 2021].
[16] Naeem, M.R.H., Gawhar, S., Adib, M.B.H., Sakib, S.A., Ahmed, A. and
Chisty, N.A.: ‘An IoT Based Smart Irrigation System’, in ‘2021 2nd
International Conference on Robotics, Electrical and Signal Processing
Techniques (ICREST)’, (2021), pp. 243–247.
[17] Cisco. Internet of Things (IoT) – The Future of IoT Miniguide: The
Burgeoning IoT Market Continues [online]. 2021. Available from:
https://www.cisco.com/c/en/us/solutions/internet-of-things/future-of-iot.
html [Accessed May 2021].
[18] Satu, M.S., Rahman, M.K., Rony, M.A., et al.: ‘COVID-19: Update,
Forecast and Assistant – An Interactive Web Portal to Provide Real-Time
Information and Forecast COVID-19 Cases in Bangladesh’, in ‘2021
International Conference on Information and Communication Technology
for Sustainable Development (ICICT4SD)’, (2021), pp. 456–460.
[19] Asif-Ur-Rahman, M., Afsana, F., Mahmud, M., et al.: ‘Toward a Heterogeneous
Mist, Fog, and Cloud-Based Framework for the Internet of Healthcare Things’
IEEE Internet of Things Journal, 2019, 6(3), pp. 4049–4062.
[20] Farhin, F., Kaiser, M.S., and Mahmud, M. ‘Towards secured service provi-
sioning for the internet of healthcare things’, in 2020 IEEE 14th
International Conference on Application of Information and Communication
Technologies (AICT) 2020 Oct 7 (pp. 1–6). IEEE.
[21] Farhin, F., Kaiser, M.S. and Mahmud, M.: ‘Secured Smart Healthcare
System: Blockchain and Bayesian Inference Based Approach’, in Kaiser, M.
S., Bandyopadhyay, A., Mahmud, M., and Ray, K. (Eds.): ‘Proceedings of
International Conference on Trends in Computational and Cognitive
Engineering’, (Springer, Berlin, 2021), pp. 455–465.
[22] Kaiser, M.S., Zenia, N., Tabassum, F., et al.: ‘6G Access Network for
Intelligent Internet of Healthcare Things: Opportunity, Challenges, and
Research Directions’, in Kaiser, M.S., Bandyopadhyay, A., Mahmud, M.,
and Ray, K. (Eds.): ‘Proceedings of International Conference on Trends in
Computational and Cognitive Engineering’, (Springer, Berlin, 2021),
pp. 317–328.
[23] Kaiser, M.S., Al Mamun, S., Mahmud, M. and Tania, M.H.: ‘Healthcare
Robots to Combat COVID-19’, in Santosh, K.C., and Joshi, A. (Eds.):
‘COVID-19: Prediction, Decision-Making, and Its Impacts’, (Springer,
Berlin, 2021), pp. 83–97.
[24] Mahmud, M. and Kaiser, M.S.: ‘Machine Learning in Fighting Pandemics:
A COVID-19 Case Study’, in Santosh, K.C., and Joshi, A. (Eds.): ‘COVID-
19: Prediction, Decision-Making, and Its Impacts’, (Springer, Berlin, 2021),
pp. 77–81.
[25] Toyama, S. and Hirayama, M.: ‘User Interface Design Method Considering
UI Device in Internet of Things System’, in ‘2018 6th International
Conference on Future Internet of Things and Cloud Workshops
(FiCloudW)’, (2018), pp. 1–6.
[26] Barcelo, M., Correa, A., Llorca, J., Tulino, A.M., Vicario, J.L. and Morell,
A.: ‘IoT-Cloud Service Optimization in Next Generation Smart
Environments’ IEEE Journal on Selected Areas in Communications, 2016,
34(12), pp. 4077–4090.
[27] Levy, E. IoT Analytics: Challenges, Applications, and Innovations [online].
2019. Available from: http://www.upsolver.com/blog/iot-analytics-chal-
lenges- applications-innovations [Accessed May 2021].
[28] Valerio, P.: ‘Is the IoT a Tech Bubble for Cities?: With More Cities Joining
the Smart City Revolution and Investing in Sensors and Other IoT Devices,
the Risk of a New Tech Bubble Is Rising’ IEEE Consumer Electronics
Magazine, 2016, 5(1), pp. 61–62.
[29] Bhuyan, H.K., Chakraborty, C., Pani, S.K. and Ravi, V.: ‘Feature and
Subfeature Selection for Classification Using Correlation Coefficient and Fuzzy
Model’ IEEE Transactions on Engineering Management, 2021, pp. 1–15.
[30] Dhawan, S., Chakraborty, C., Frnda, J., Gupta, R., Rana, A.K. and Pani, S.K.:
‘SSII: Secured and High-Quality Steganography Using Intelligent Hybrid
Optimization Algorithms for IoT’ IEEE Access, 2021, 9, pp. 87563–87578.
[31] Frustaci, M., Pace, P., Aloi, G. and Fortino, G.: ‘Evaluating Critical Security
Issues of the IoT World: Present and Future Challenges’ IEEE Internet of
Things Journal, 2018, 5(4), pp. 2483–2495.
[32] Dominicbetts. What is Azure IoT Central [online]. 2021. Available from:
https://docs.microsoft.com/en-us/azure/iot- central/core/overview-iot-central
[Accessed May 2021].
[33] Nofer, M., Gomber, P., Hinz, O. and Schiereck, D.: ‘Blockchain’ Business &
Information Systems Engineering, 2017, 59(3), pp. 183–187.
[34] Zheng, Z., Xie, S., Dai, H., Chen, X. and Wang, H.: ‘An Overview of
Blockchain Technology: Architecture, Consensus, and Future Trends’, in
‘2017 IEEE International Congress on Big Data (BigData Congress)’,
(2017), pp. 557–564.
[35] Chowdhury, S.M.H.M., Jahan, F., Sara, S.M. and Nandi, D.: ‘Secured
Blockchain Based Decentralised Internet: A Proposed New Internet’, in
‘Proceedings of the International Conference on Computing Advancements’,
(Association for Computing Machinery, New York, 2020), pp. 1–7.
[36] MarketsandMarkets. Blockchain Market by Component (Platform and

Services), Provider (Application, Middleware, and Infrastructure), Type
(Private, Public, and Hybrid), Organization Size, Application Area (BFSI,
Government, IT & Telecom), and Region – Global Forecast to 2025 [online].
2020. Available from: https://www.marketsandmarkets.com/Market-Reports/
blockchain- technology-market-90100890.html [Accessed May 2021].
[37] Sayeed, S. and Marco-Gisbert, H.: ‘Assessing Blockchain Consensus and Security
Mechanisms Against the 51% Attack’ Applied Sciences, 2019, 9(9), p. 1788.
[38] Onik, M.M.H., Aich, S., Yang, J., Kim, C.-S. and Kim, H.-C.: ‘Chapter 8 –
Blockchain in Healthcare: Challenges and Solutions’, in Dey, N., Das, H.,
Naik, B., and Behera, H.S. (Eds.): ‘Big Data Analytics for Intelligent
Healthcare Management’, (Academic Press, New York, 2019), pp. 197–226.
[39] Hassija, V., Chamola, V., Saxena, V., Jain, D., Goyal, P. and Sikdar, B.: ‘A
Survey on IoT Security: Application Areas, Security Threats, and Solution
Architectures’ IEEE Access, 2019, 7, pp. 82721–82743.
[40] Yang, J., Onik, M.M.H., Lee, N.-Y., Ahmed, M. and Kim, C.-S.: ‘Proof-of-
Familiarity: A Privacy-Preserved Blockchain Scheme for Collaborative
Medical Decision-Making’ Applied Sciences, 2019, 9(7), p. 1370.
[41] Zhang, S. and Lee, J.-H.: ‘Analysis of the Main Consensus Protocols of
Blockchain’ ICT Express, 2020, 6(2), pp. 93–97.
[42] Kaur, M., Khan, M.Z., Gupta, S., Noorwali, A., Chakraborty, C. and Pani, S.
K.: ‘MBCP: Performance Analysis of Large Scale Mainstream Blockchain
Consensus Protocols’ IEEE Access, 2021, 9, pp. 80931–80944.
[43] Niranjanamurthy, M., Nithya, B.N. and Jagannatha, S.: ‘Analysis of
Blockchain Technology: Pros, Cons and SWOT’ Cluster Computing, 2019,
22(6), pp. 14743–14757.
[44] Choi, S. and, Lee, J.-H.: ‘Blockchain-Based Distributed Firmware Update
Architecture for IoT Devices’ IEEE Access, 2020, 8, pp. 37518–37525.
[45] Czachorowski, K., Solesvik, M. and Kondratenko, Y.: ‘The Application of
Blockchain Technology in the Maritime Industry’, in Kharchenko, V.,
Kondratenko, Y., and Kacprzyk, J. (Eds.): ‘Green IT Engineering: Social,
Business and Industrial Applications’, (Springer International Publishing,
Berlin, 2019), pp. 561–577.
[46] Rhydwan, A.H.M., Choyon, M.M.S., Iftekhar, M.N., Hasan Sad, A.M. and
Nandi, D.: ‘Smart Grid Implementation With Consortium Blockchain: A
Proposed Model for Bangladesh’, in ‘2020 Emerging Technology in
Computing, Communication and Electronics (ETCCE)’, (2020), pp. 1–6.
[47] Rodrigues, J.J.P.C., Segundo, D.B.D.R., Junqueira, H.A., et al.: ‘Enabling
Technologies for the Internet of Health Things’ IEEE Access, 2018, 6,
pp. 13129–13141.
[48] Ye, J.: ‘The Role of Health Technology and Informatics in a Global Public
Health Emergency: Practices and Implications From the COVID-19
Pandemic’ JMIR Medical Informatics, 2020, 8(7), p. e19866.
[49] Faria, T.H., Shamim Kaiser, M., Hossian, C.A., Mahmud, M., Al Mamun, S.
and Chakraborty, C.: ‘Smart City Technologies for Next Generation
Healthcare’, in Chakraborty, C., Lin, J.C.-W., and Alazab, M. (Eds.): ‘Data-

Driven Mining, Learning and Analytics for Secured Smart Cities: Trends and
Advances’, (Springer International Publishing, Berlin, 2021), pp. 253–274.
[50] Sad, A.S.M.M.H., Choyon, M.M.S., Rhydwan, A.H.M., Shikder, K. and
Hossain, C.A.: ‘Nekray: A Linux Kernel-Based Customized Operating
System for Information Kiosk’ IJERTCS, 2021, 12(3), pp. 49–68.
[51] Choyon, M.M.S., Rahman, M., Kabir, M.M. and Mridha, M.F.: ‘IoT Based
Health Monitoring Automated Predictive System to Confront COVID-19’,
in ‘2020 IEEE 17th International Conference on Smart Communities:
Improving Quality of Life Using ICT, IoT and AI (HONET)’, (2020),
pp. 189–193.
[52] Michalska, A.: ‘Could Your Fitbit or Apple Watch Detect Early COVID-
19 Symptoms?’ Reuters, 2020.
[53] Miller, D.J., Capodilupo, J.V., Lastella, M., et al.: ‘Analyzing Changes in
Respiratory Rate to Predict The Risk of COVID-19 Infection’ medRxiv,
2020, 2020.06.18.20131417.
[54] Rahman, M.A. and Hossain, M.S.: ‘An Internet of Medical Things-Enabled
Edge Computing Framework for Tackling COVID-19’ IEEE Internet of
Things Journal, 2021, 8, pp. 15847–15854.
[55] Bai, L., Yang, D., Wang, X., et al.: ‘‘Chinese Experts’ Consensus on the
Internet of Things-Aided Diagnosis and Treatment of Coronavirus Disease
2019 (COVID-19)’ Clinical eHealth, 2020, 3, pp. 7–15.
[56] Ruiz-Fernández, D., Marcos-Jorquera, D., Gilart-Iglesias, V., Vives-Boix, V.
and Ramı́rez-Navarro, J.: ‘Empowerment of Patients With Hypertension
through BPM, IoT and Remote Sensing’ Sensors, 2017, 17(10), p. 2273.
[57] Lomotey, R.K., Pry, J. and Sriramoju, S.: ‘Wearable IoT Data Stream
Traceability in a Distributed Health Information System’ Pervasive and
Mobile Computing, 2017, 40, pp. 692–707.
[58] Taiwo, O. and Ezugwu, A.E.: ‘Smart Healthcare Support for Remote Patient
Monitoring During Covid-19 Quarantine’ Informatics in Medicine
Unlocked, 2020, 20, p. 100428.
[59] Vedaei, S.S., Fotovvat, A., Mohebbian, M.R., et al.: ‘COVID-SAFE: An
IoT-Based System for Automated Health Monitoring and Surveillance in
Post-Pandemic Life’ IEEE Access, 2020, 8, pp. 188538–188551.
[60] Li, S.-H., Lin, B.-S., Tsai, C.-H., Yang, C.-T. and Lin, B.-S.: ‘Design of
Wearable Breathing Sound Monitoring System for Real-Time Wheeze
Detection’ Sensors (Basel), 2017, 17(1).
[61] de Morais Barroca Filho, I., Aquino, G., Malaquias, R.S., Girão, G. and
Melo, S.R.M.: ‘An IoT-Based Healthcare Platform for Patients in ICU Beds
During the COVID-19 Outbreak’ IEEE Access, 2021, 9, pp. 27262–27277.
[62] Winn, Z. Real-Time Data for a Better Response to Disease Outbreaks
[online]. 2020. Available from: https://news. mit.edu/2020/kinsa-health-
0821 [Accessed April 2021].
[63] Fernández-Caramés, T.M., Froiz-Mı́guez, I. and Fraga-Lamas, P.: ‘An IoT
and Blockchain Based System for Monitoring and Tracking Real-Time
Occupancy for COVID-19 Public Safety’ Engineering Proceedings, 2020,

2(1), p. 67.
[64] Marr, B. Robots And Drones Are Now Used To Fight COVID-19 [online]. 2020.
Available from: https:// www.forbes.com/sites/bernardmarr/2020/03/18/how-
robots-and- drones-arehelping-to-fight-coronavirus/ [Accessed April 2021].
[65] Gascueña, D. Drones to Stop the COVID-19 Epidemic [online]. 2020.
Available from: https://www.bbva.com/en/drones-to-stop-the-covid-19-epi-
demic/ [Accessed April 2021].
[66] Rosebrock, A. OpenCV Social Distancing Detector [online]. 2020. Available
from: https://www.pyimagesearch.com/2020/06/01/opencv-social-distan-
cing-detector/ [Accessed April 2021].
[67] Abdulrazaq, M., Zuhriyah, H., Al-Zubaidi, S., Karim, S., Ramli, R. and
Yusuf, E.: ‘NOVEL COVID-19 Detection And Diagnosis System Using IoT
Based Smart Helmet’ International Journal of Psychosocial Rehabilitation,
2020, 24, pp. 2296–2303.
[68] Abdulrazaq, M., Istiqomah, N., Zuhriyah, H., et al.: ‘2019 Novel
Coronavirus Disease (Covid-19): Detection and Diagnosis System Using IoT
Based Smart Glasses’ 2020, 29.
[69] Jung, Y. and Agulto, R.: ‘A Public Platform for Virtual IoT-Based
Monitoring and Tracking of COVID-19’ Electronics, 2021, 10(1), p. 12.
[70] Garg, C., Bansal, A. and Padappayil, R.P.: ‘COVID-19: Prolonged Social
Distancing Implementation Strategy Using Blockchain-Based Movement
Passes’ Journal of Medical Systems, 2020, 44(9), p. 165.
[71] Garg, L., Chukwu, E., Nasser, N., Chakraborty, C. and Garg, G.:
‘Anonymity Preserving IoT-Based COVID-19 and Other Infectious Disease
Contact Tracing Model’ IEEE Access, 2020, 8, pp. 159402–159414.
[72] Tedeschi, P., Bakiras, S. and Di Pietro, R.: ‘IoTrace: A Flexible, Efficient,
and Privacy-Preserving IoT-enabled Architecture for Contact Tracing’
arXiv:2007.11928 [cs], 2021.
[73] Kolhar, M., Al-Turjman, F., Alameen, A. and Abualhaj, M.M.: ‘A Three
Layered Decentralized IoT Biometric Architecture for City Lockdown
During COVID-19 Outbreak’ IEEE Access, 2020, 8, pp. 163608–163617.
[74] Yang, G.-Z., J. Nelson, B., Murphy, R.R., et al.: ‘Combating COVID-19—
The Role of Robotics in Managing Public Health and Infectious Diseases’
Science Robotics, 2020, 5(40), p. eabb5589.
[75] Xenex. Xenex LightStrike Robot – Destroy Deadly Pathogens and
Deactivate SARS-CoV-2 [Online]. 2021. Available from: https://xenex.com/
[Accessed April 2021].
[76] Ackerman, E. Autonomous Robots Are Helping Kill Coronavirus in Hospitals
[online]. 2020. Available from: https://spectrum.ieee.org/automaton/robotics/
medical-robots/autonomous-robots-are-helping-kill-coronavirus-in-hospitals
[77] Sharma, M. How Drones Are Being Used to Combat COVID-19 [online].
2020. Available from: https://www.geospatialworld.net/blogs/how-drones-
are-being-used- to-combat-covid19/ [Accessed April 2021].
[78] Hossain, C., Mohamed, M.A., Zishan, M., Ahasan, R. and Sharun, S.:
‘Awareness on e-Health Among Undergraduate Students in Bangladesh’
Indian Journal of Public Health Research & Development, 2019, 10, p. 636.
[79] Medicaaid. Telemedicine [online]. 2020. Available from: https://www.med-
icaid.gov/medicaid/benefits/ telemedicine/index.html [Accessed April
2021].
[80] Matsos, S. In Fight Against Coronavirus, Telemedicine Ramps Up at Johns
Hopkins [online]. 2020. Available from: https://www.hopkinsmedicine.org/
coronavirus/articles/telemedicine.html [Accessed April 2021].
[81] Athena, P., Rumsfeld, J.S. and Wessler, J.D.: ‘Telehealth Is Having a
Moment’ Journal of the American College of Cardiology, 2020, 75(23),
pp. 2989–2991.
[82] Gilgore, S. GWU Hospital tackles COVID-19 with new testing site, tele-
medicine and outreach on D.C.’s east side [online]. 2020. Available from:
https://www.bizjournals.com/washington/news/2020/04/08/gwu- hospital-
tackles-covid-19-with-new-testing.html [Accessed April 2021].
[83] Shah, M. Industry Voices—How Rush University Medical Center’s Virtual
Investments Became Central to Its COVID-19 Response [online]. Available
from: https://www. fiercehealthcare.com/hospitals-health-systems/industry-
voices- how-rushuniversity-system-for-health-s-virtual [Accessed April
2021].
[84] WAS. WAS ambulance and special vehicles of the WAS brand [online].
2021. Available from: https://www.was- vehicles.com/en/home.html
[85] Hossain, C., Zishan, M. and Ahasan, R.: ‘A Review on the Security Issues of
Telemedicine Network’ IJIREEICE, 2014, pp. 2183–218.
[86] Siriwardhana, Y., De Alwis, C., Gür, G., Ylianttila, M. and Liyanage, M.:
‘The Fight Against the COVID-19 Pandemic with 5G Technologies’ IEEE
Engineering Management Review, 2020, 48(3), pp. 72–84.
[87] Dawkins, T. How COVID-19 Could Open the Door for Driverless [online].
2020. Available from: https://www.weforum.org/agenda/2020/04/how-
covid-19-could-open-the-door-for-driverless-deliveries/ [Accessed April
2021].
[88] Končar, J., Grubor, A., Marić, R., Vučenović, S. and Vukmirović, G.: ‘Setbacks
to IoT Implementation in the Function of FMCG Supply Chain Sustainability
During COVID-19 Pandemic’ Sustainability, 2020, 12(18), p. 7391.
[89] Rivest, R.L., Weitzner, D.J., Ivers, L.C., Soibelman, I. and Zissman, M.A.:
‘PACT: Private Automated Contact Tracing’ 2020, p. 3.
[90] Song, J., Gu, T., Feng, X., Ge, Y. and Mohapatra, P.: ‘Blockchain Meets
COVID-19: A Framework for Contact Information Sharing and Risk
Notification System’ arXiv preprint arXiv:2007.10529, 2020.
[91] Xu, H., Zhang, L., Onireti, O., Fang, Y., Buchanan, W.J. and Imran, M.A.:
‘BeepTrace: Blockchain-Enabled Privacy-Preserving Contact Tracing for
COVID-19 Pandemic and Beyond’ IEEE Internet of Things Journal, 2020,
8, pp. 3915–3929.
[92] Klaine, P.V., Zhang, L., Zhou, B., Sun, Y., Xu, H. and Imran, M.: ‘Privacy-
Preserving Contact Tracing and Public Risk Assessment Using Blockchain for
COVID-19 Pandemic’ IEEE Internet of Things Magazine, 2020, 3(3), pp. 58–63.
[93] Torky, M. and Hassanien, A.E.: ‘COVID-19 Blockchain Framework:
Innovative Approach’ arXiv preprint arXiv:2004.06081, 2020.
[94] Gupta, A, Chinmay, C. and Gupta, B: ‘Secure Transmission of EEG Data Using
Watermarking Algorithm for the Detection of Epileptical Seizures’ Traitement
du Signal, 2021, 38(2), pp. 473–479, https://doi.org/10.18280/ts.380227.
[95] Azim, A., Islam, M.N. and Spranger, P.E.: ‘Blockchain and Novel
Coronavirus: Towards Preventing COVID-19 and Future Pandemics’
Iberoamerican Journal of Medicine, 2020, 2(3), pp. 215–218.
[96] Manoj, M., Srivastava, G., Somayaji, S.R.K., Gadekallu, T.R., Maddikunta,
P.K.R. and Bhattacharya, S.: ‘An Incentive Based Approach for COVID-19
Planning Using Blockchain Technology’, in ‘2020 IEEE Globecom
Workshops (GC Wkshps)’, (2020), pp. 1–6.
[97] Choudhury, H., Goswami, B. and Gurung, S.K.: ‘CovidChain: An
Anonymity Preserving Blockchain Based Framework for Protection
Against Covid-19’ ArXiv, 2020.
[98] Hasan, H.R., Salah, K., Jayaraman, R., et al.: ‘Blockchain-Based Solution
for COVID-19 Digital Medical Passports and Immunity Certificates’ IEEE
Access, 2020, 8, pp. 222093–222108.
[99] Resiere, D., Resiere, D. and Kallel, H.: ‘Implementation of Medical and
Scientific Cooperation in the Caribbean Using Blockchain Technology in
Coronavirus (Covid-19) Pandemics’ Journal of Medical Systems, 2020, 44(7).
[100] Rab, S., Javaid, M., Haleem, A. and Vaishya, R.: ‘Face Masks Are New
Normal After COVID-19 Pandemic’ Diabetes & Metabolic Syndrome:
Clinical Research & Reviews, 2020, 14(6), pp. 1617–1619.
[101] Ahir, S., Telavane, D. and Thomas, R.: ‘The Impact of Artificial
Intelligence, Blockchain, Big Data and Evolving Technologies in
Coronavirus Disease – 2019 (COVID-19) Curtailment’, in ‘2020
International Conference on Smart Electronics and Communication
(ICOSEC)’, (2020), pp. 113–120.
[102] Christodoulou, K., Christodoulou, P., Zinonos, Z., Carayannis, E.G. and
Chatzichristofis, S.A.: ‘Health Information Exchange With Blockchain
amid Covid-19-like Pandemics’, in ‘2020 16th International Conference on
Distributed Computing in Sensor Systems (DCOSS)’, (2020), pp. 412–417.
[103] Khatoon, A. Use of Blockchain Technology to Curb Novel Coronavirus
Disease (COVID-19) Transmission. Preprints 2020, 2020050117 (doi:
10.20944/preprints202005.0117.v1).
[104] Alam T. ‘Internet of things and blockchain-based framework for
Coronavirus (COVID-19) disease’. Available at SSRN: http://dx.doi.org/
10.2139/ssrn.3660503.
[105] Herland, M., Khoshgoftaar, T.M. and Bauder, R.A.: ‘Big Data Fraud
Detection Using Multiple Medicare Data Sources’ Journal of Big Data,
2018, 5(1), p. 29.
[106] Batool, A., Toyama, K., Veinot, T., Fatima, B. and Naseem, M.: ‘Detecting
Data Falsification by Front-line Development Workers: A Case Study of
Vaccination in Pakistan’, in ‘Proceedings of the 2021 CHI Conference on
Human Factors in Computing Systems’, (Association for Computing
Machinery, New York, 2021), pp. 1–14.
[107] Marbouh, D., Abbasi, T., Maasmi, F., et al.: ‘Blockchain for COVID-19:
Review, Opportunities, and a Trusted Tracking System’ Arabian Journal
for Science and Engineering, 2020, 45(12), pp. 9895–9911.
[108] Ahmad, R.W., Salah, K., Jayaraman, R., Yaqoob, I., Ellahham, S., and
Omar, M. ‘Blockchain and COVID-19 pandemic: Applications and chal-
lenges’, IEEE TechRxiv. 2020 Sep 17.
[109] Ramachandran, A. and Kantarcioglu, M.: ‘SmartProvenance: A Distributed,
Blockchain Based DataProvenance System’, in ‘Proceedings of the Eighth
ACM Conference on Data and Application Security and Privacy’,
(Association for Computing Machinery, New York, 2018), pp. 35–42.
[110] Siva Rama Krishnan, S., Manoj, M.K., Gadekallu, T.R., et al.: ‘A Blockchain-
Based Credibility Scoring Framework for Electronic Medical Records’, in
‘2020 IEEE Globecom Workshops (GC Wkshps)’, (2020), pp. 1–6.
[111] Sharma, B., Sekharan, C.N. and Zuo, F.: ‘Merkle-Tree Based Approach for
Ensuring Integrity of Electronic Medical Records’, in ‘2018 9th IEEE
Annual Ubiquitous Computing, Electronics Mobile Communication
Conference (UEMCON)’, (2018), pp. 983–987.
[112] Boudagdigue, C., Benslimane, A., Kobbane, A. and Liu, J.: ‘Trust
Management in Industrial Internet of Things’ IEEE Transactions on
Information Forensics and Security, 2020, 15, pp. 3667–3682.
[113] Dai, H.-N., Imran, M. and Haider, N.: ‘Blockchain-Enabled Internet of
Medical Things to Combat COVID-19’ IEEE Internet of Things Magazine,
2020, 3(3), pp. 52–57.
[114] Aileni, R.M. and Suciu, G., 2020. ‘IoMT: A blockchain perspective’, in
Decentralised Internet of Things (pp. 199–215). Springer, Cham.
[115] Srivastava, G., Crichigno, J. and Dhar, S.: ‘A Light and Secure Healthcare
Blockchain for IoT Medical Devices’, in ‘2019 IEEE Canadian Conference
of Electrical and Computer Engineering (CCECE)’, (2019), pp. 1–5.
[116] Celesti, A., Ruggeri, A., Fazio, M., Galletta, A., Villari, M. and Romano,
A.: ‘Blockchain-Based Healthcare Workflow for Tele-Medical Laboratory
in Federated Hospital IoT Clouds’ Sensors, 2020, 20(9), p. 2590.
[117] Košt’ál, K., Helebrandt, P., Belluš, M., Ries, M. and Kotuliak, I.:
‘Management and Monitoring of IoT Devices Using Blockchain’ Sensors,
2019, 19(4), p. 856.
[118] Ahmad, R.W., Salah, K., Jayaraman, R., Yaqoob, I., Ellahham, S. and Omar,
M.: ‘The Role of Blockchain Technology in Telehealth and Telemedicine’
International Journal of Medical Informatics, 2021, 148, p. 104399.
[119] Kim, D., Choi, J. and Han, K.: ‘Risk Management-Based Security
Evaluation Model for Telemedicine Systems’ BMC Medical Informatics
and Decision Making, 2020, 20(1), p. 106.
[120] Kordestani, H., Barkaoui, K. and Zahran, W.: ‘HapiChain: A Blockchain-

Based Framework for Patient-Centric Telemedicine’, in ‘2020 IEEE 8th
International Conference on Serious Games and Applications for Health
(SeGAH)’, (2020), pp. 1–6.
[121] Guo, R., Shi, H., Zheng, D., Jing, C., Zhuang, C. and Wang, Z.: ‘Flexible
and Efficient Blockchain-Based ABE Scheme With Multi-Authority for
Medical on Demand in Telemedicine System’ IEEE Access, 2019, 7,
pp. 88012–88025.
[122] Tzagkarakis, C., Petroulakis, N. and Ioannidis, S.: ‘Botnet Attack Detection
at the IoT Edge Based on Sparse Representation’, in ‘2019 Global IoT
Summit (GIoTS)’, (2019), pp. 1–6.
[123] De Rango, F., Tropea, M. and Fazio, P.: ‘Mitigating DoS Attacks in IoT
EDGE Layer to Preserve QoS Topics and Nodes’ Energy’, in ‘IEEE
INFOCOM 2020 – IEEE Conference on Computer Communications
Workshops (INFOCOM WKSHPS)’, (2020), pp. 842–847.
[124] Luo, X., Yan, Q., Wang, M. and Huang, W.: ‘Using MTD and SDN-
Based Honeypots to Defend DDoS Attacks in IoT’, in ‘2019
Computing, Communications and IoT Applications (ComComAp)’,
(2019), pp. 392–395.
[125] Liu, G., Quan, W., Cheng, N., et al.: ‘Softwarized IoT Network Immunity
Against Eavesdropping With Programmable Data Planes’ IEEE Internet of
Things Journal, 2021, 8(8), pp. 6578–6590.
[126] Clark, G.W., Doran, M.V. and Andel, T.R.: ‘Cybersecurity issues in
robotics’, in ‘2017 IEEE Conference on Cognitive and Computational
Aspects of Situation Management (CogSIMA)’, (2017), pp. 1–5.
[127] Taylor, C. Robots Could Take Over 20 Million Jobs by 2030, Study Claims
[online]. 2019. Available https://www.cnbc.com/2019/06/26/robots-could-
take-over-20- million-jobs-by-2030-study-claims.html [Accessed May 2021].
[128] ‘IoT Standardization and Implementation Challenges – IEEE Internet of
Things’, https://iot.ieee.org/newsletter/july-2016/iot-standardization-and-
implementation-challenges.html, accessed May 2021.
[129] Mbunge, E., Akinnuwesi, B., Fashoto, S.G., Metfula, A.S. and Mashwama, P.:
‘A Critical Review of Emerging Technologies for Tackling COVID-19
Pandemic’ Human Behavior and Emerging Technologies, 2021, 3(1), pp. 25–39.
[130] Alrashdi, I., Alqazzaz, A., Aloufi, E., Alharthi, R., Zohdy, M. and Ming, H.:
‘AD-IoT: Anomaly Detection of IoT Cyberattacks in Smart City Using
Machine Learning’, in ‘2019 IEEE 9th Annual Computing and
Communication Workshop and Conference (CCWC)’, (2019), pp. 0305–0310.
[131] Zarca, A.M., Bernabe, J.B., Skarmeta, A. and Alcaraz Calero, J.M.:
‘Virtual IoT HoneyNets to Mitigate Cyberattacks in SDN/NFV-Enabled
IoT Networks’ IEEE Journal on Selected Areas in Communications, 2020,
38(6), pp. 1262–1277.
[132] Park, M., Oh, H. and Lee, K.: ‘Security Risk Measurement for Information
Leakage in IoT-Based Smart Homes from a Situational Awareness
Perspective’ Sensors, 2019, 19(9), p. 2148.
[133] Sharma, M., Samriti and Singh, G.: ‘Need and Design of Smart and Secure
Energy-Efficient IoT-Based Healthcare Framework’, in Mittal, M.,
Tanwar, S., Agarwal, B., and Goyal, L.M. (Eds.): ‘Energy Conservation for
IoT Devices: Concepts, Paradigms and Solutions’, (Springer, Berlin, 2019),
pp. 259–281.
[134] Ghiasi, M., Dehghani, M., Niknam, T., Kavousi-Fard, A., Siano, P. and
Alhelou, H.H.: ‘Cyber-Attack Detection and Cyber-Security
Enhancement in Smart DC-Microgrid Based on Blockchain
Technology and Hilbert Huang Transform’ IEEE Access, 2021, 9,
pp. 29429–29440.
[135] Singh, S.K., Salim, M.M., Cho, M., Cha, J., Pan, Y. and Park, J.H.: ‘Smart
Contract-Based Pool Hopping Attack Prevention for Blockchain Networks’
Symmetry, 2019, 11(7), p. 941.
[136] Salimitari, M., Chatterjee, M. and Fallah, Y.P.: ‘A Survey on Consensus
Methods in Blockchain for Resource-Constrained IoT Networks’ Internet
of Things, 2020, 11, p. 100212.
Chapter 2
Application of big data and computational
intelligence in fighting COVID-19 epidemic
Joseph Bamidele Awotunde1, Chinmay Chakraborty2 and
Gbemisola Janet Ajamu3
Abstract
The current epidemic called coronavirus (COVID-19) is wreaking havoc on
society, humanity, and causing economic difficulties around the world. Many
techniques have been attempted to manage and contain the COVID-19 outbreak;
however, many governments remain powerless to combat and contain the virus.
Big data is driving the digital revolution in an increasingly knowledge-driven,
healthcare-innovation-driven, and connected society. The combination of com-
putational intelligence (CI) and big data analytics (BDAs) has developed methods
that make accessing and processing vast amounts of data easier and less
demanding on human expert. Hence, in combating the outbreak, big data and CI
can be applied since the use of both technologies empowered BDA and yielded
imaginable results in combating infectious diseases globally. Therefore, this
chapter reviews the applicability and importance of big data and CI methods to
data produced from the countless of ubiquitously connected healthcare devices
that produced entrenched and distributed information handling capabilities in
fighting COVID-19 outbreak. The use of CI in BDAs has resulted in knowledge-
based system that transformed big data into big knowledge with new approaches
and visions in order to provide people with better understanding and information-
driven results. There have been tremendous positive results using IoT-based
capture data with BDA and CI models for monitoring, diagnosis, and prediction
of COVID-19 outbreak. The huge amount of data can be managed using CI and
BDA by developing models that will reduce the spread of any infectious diseases
to monitoring, tracking and in the production of drugs and vaccines for the
treatment of the any outbreak globally.
1
Department of Computer Science, University of Ilorin, Ilorin, Nigeria
2
Department of Electronics and Communication Engineering, Birla Institute of Technology, Ranchi,
India
3
Department Agricultural Extension and Rural Development, Landmark University, Omu-Aran, Nigeria
Keywords: Big disease diagnosis and prediction; COVID-19 outbreak;

Healthcare system; Big data analytics; Computational intelligence; Big data;
Diseases diagnosis
2.1 Introduction
The coronavirus (COVID-19) epidemic outbreak, which began in Wuhan, China in

December 2019, has already spread around the globe. According to the WHO, in
213 countries/regions of the world as at 10:35 AM CEST, there were 199,203,664
COVID-19 cases worldwide as of August 2, 2021, with 4,244,184 global deaths [1].
Although there have been new vaccine provided for citizens around the world, but
the drugs are not enough considering the population globally, and the fear of side
effects have made many to reject the vaccine available. Hence, there is need for a
better way of managing this pandemic globally. Also, the community-based control
strategies adopted globally have not proved useful to curtail the outbreak [2,3].
The cause of COVID-19 has globally affected almost the aspect of human
activities because it is highly contagious. Every nation adopts the system of con-
tainment, monitoring, and lockdown to be able to curtail the spread of the outbreak
by isolating those that have infected the communities in order for them not to
spread the pandemic within the community further. Patients with the virus may
have modest symptoms like sneezing, sore throat, and fever or major symptoms
that can lead to death like heart failure and pneumonia [3]. COVID-19 is the cor-
onavirus induced by the SARS-CoV-2 transmittable condition. Thanks to recent
advances in analytical techniques brought about by information and communica-
tion technologies (ICTs), big data analytics (BDAs), and computational intelligence
(CI) can be used to manage the enormous and unprecedented volume of data
generated by medical data. These huge data can be used to monitor patients’ in real
time, giving daily reports of the situation of any outbreak, tracking of the pandemic,
daily situation briefings, and public updates. According to a recent assessment,
about 86.60% of all patients infected so far are susceptible to the COVID-19 pan-
demic given a median age of around 47 years [4].
In this larger proclivity outbreak, men are taking the lead, while studies have
not found a substantial gender difference [5]. The primary carrier of COVID-19
outbreak is through human-to-human transmission having aerosols, respiratory
aspirates, droplets, feces, and direct exposure all being classified as carriers by
China’s National Health Commission [6]. Although there is no substantial proof of
infrequent instances of SARS-CoV-2 being transmitted vertically, it has been
shown that this could be a route for COVID-19 to spread [7].
The various integrations of devices in healthcare sectors have generated big
data, and the advancement of CI in the processes of this huge data has achieved
significant success in several infectious diseases like influenzas [8], SARS [9], and
MERS [10]. The CI models have been used to predict and diagnose various
infectious diseases, thus, help in dealing with their several challenges. This has
Application of big data and computational intelligence 35
result to the development of various CI systems for fighting COVID-19 outbreak.

The CI system plays a crucial role for combating COVID-19 pandemic by building
robust system used to diagnose, monitor, forecast, and develop COVID-19 vaccine.
The application of CI would require various types of CI models like computational
modeling, artificial intelligence (AI), data analytics, and particularly machine
learning, a powerful subfield of CI. The integration of various devices with CI
mechanisms in developing different applications would require various models to
be able to fight any infectious diseases. Researchers have made efforts to develop
systems and applications using CI models for the fight against COVID-19 outbreak.
The use of CI and analytical methodologies to predict unpredictable or fore-
seeable results is known as predictive analysis [10,11]. It answers the question,
“What’s the next step?” It also uses historical and current data to forecast future
behaviors, patterns, and activities. To make a better prediction model, automated
CI and statistical analysis with quantitative questions play prominent role [11].
Predictive models must be built by experts for forecasting in predictive analytics
[11]. The CI is a catch-all term for a variety of information and control tools. The
generated data from IoT-based wearable devices will be worthless without the use
of CI models to generate useful insight that can be useful for both physician and the
patients. Wearable app creators are gradually integrating CI engines into wearable
medical apps for this purpose. Furthermore, the integration of CI models has really
helped in the development of a better healthcare system connected to various IoT
devices, smartphones, and wearable sensors to collect data that will improve the
understanding of patient problems by the physicians [12,13].
The recent technological development with innovations in data sciences has
facilitated the gathering of a huge amount of data, especially in healthcare sys-
tems, thus termed big data, and clouding has been the warehouses where such
huge data can be safely stored [14,15]. Researchers must broaden researches
beyond the standard surveillance model in order to make an efficient use of the
captured data. The research should be based on the models that will make use of
huge data to generate various results that are focusing on treatment-oriented
healthcare systems by narrowing the BDA to diagnose, track, monitor, and treat
various diseases. As a result, big data can be used in the heterogeneous population
for various purposes to reflect the needs of each sector for a better decision-
making [16]. This necessities speedy research that can make use of modern
computing technologies to process the huge data gathered in healthcare systems
and make use appropriate tools to lessen their implantation complexity [17]. Big
data generated in healthcare systems can be used for proper diagnosis and treat-
ment of patients with given accurate results.
The mutual collaboration in healthcare industries has resulted in the collection
of a huge amount of data from various sources [18]. This collaboration has brought
medical workers from various departments to achieve a common goal of reducing
medical cost, and errors as a result inaccuracy in medical diagnosis. A huge amount
of data generated from each section of the hospitals from various departments like
nurses, pathologists, radiologists, and laboratory technologists is used. In addition
to deliver high-quality and standardized healthcare services, data is obtained from a
variety of sources by various stakeholders, which can be used for monitoring,

prediction, diagnosis, and treatment of various diseases.
As a result, sophisticated technologies such as virtualization and cloud com-
puting can be used in the healthcare system to process large amounts of data effi-
ciently. These make the healthcare system as big data-generated industries, which
have been developed in recent years. Meanwhile, advancements in ICT have resulted
in the development of huge data from various sources in recent years. Big data is
altering every element of human life in the twenty-first century, including biology
and medicine [19]. The introduction of electronic medical records to replace the
paper medical records transformed the healthcare systems into big data-generated
sectors and resulted into massive data explosion [20]. Thus, it helps to make
evidence-based decisions that ultimately improve patient diagnosis, monitoring, and
treatments by doctors, policy-makers, and the epidemiologists [21]. Big data has
been an imperative force that must be properly utilized and grasped but not only a
modern reality for biomedical scientists in the pursuit of new knowledge [22].
As a result, this chapter addresses how CI and big data could be used to handle
and combat COVID-19 pandemic by improving traditional public health ways to
regulate, monitor, and detect infected people, hence reducing the health impact of
the outbreak. The chapter is arranged as follows: Section 2.2 presents the applica-
tions of CI in modern days in helping the reduction of COVID-19 outbreak.
Section 2.3 presents applications of BDAs during this pandemic. Section 2.4 dis-
cusses the problems and challenges of using CI and BDA in fighting the infectious
outbreak. Section 2.5 presents the practical applications of CI during COVID-19
pandemic for the classification of the disease and finally the conclusion is presented
in Section 2.6.
2.2 Applicability of computational intelligence in

combating COVID-19 pandemic
The technological innovations have really helped the healthcare professionals all
through this truculence health challenges to monitor the spread of COVID-19
pandemic. CI with enabled BDAs can be used for real-time monitoring of infected
patients and helped to classify high-risk patients, which has helped to promptly
monitor the spread of this pandemic. Based on thorough examination, CI can use to
predict mortality risk by the physicians. Through mass screening, medical support,
warnings, and suggestions about how to manage diseases, CI enabled with BDAs
aids in the fight against epidemics [23,24]. As a dependable medical procedure, this
innovation can be used to produce drugs and vaccines to treat infected patients and
be used to report the performance of these intervention programs.
In 1994, Bezdek formally defined that CI [25,26] has a computationally
intelligent system if it deals with data at basic level like in the case of image pixels
and does not use prior information that have been generated. Pattern recognition is
one of such examples of such computational intelligent system that did not use
prior information to generate result using CI model. AI is one of the branches of CI
according to Bezdek’s description. In the real sense, both AI and CI target the rea-
lization of general intelligence in real-world models. The authors in [27] explained
the distinction in both AI and CI by suggesting that hard computing technologies are
made of the former, whereas soft computing technologies are made of the latter.
We may therefore assume that there are two forms of machine intelligence: (1)
CI, defined by the hard-computing principle, and (2) AI developed by the soft-
computing theory. Compared to hard-computing-based CI, the advantages of the
soft-computing concept allow AI to adapt to several different conditions. The hard
computing models are built using Boolean logic that is based on true or false
values. The main challenge of this method is that it cannot be interpreted easily in
natural language problems. The ambiguous cases are fully taken care of with soft
computing models using fuzzy logic. One patented feature of CI is this form of
logic that approximated the human brain through the aggregation of data into
partial facts [27].
CI focuses on problems that, in theory, only individuals and animals can solve
and require intellect. It is a branch of computer science without operational com-
putational algorithms, which involve only intelligence to study and solve any
challenges. It studies problems for which no operational computational algorithms
exist. The CI allows the addition of new ideas through time and serves as patronage
of other existing algorithms like AI [27–29]. The use of CI methods has resulted in
a higher level of focus in the research community. Machine learning algorithms, as
stated in a number of recent studies, provide superior detection accuracy when
compared to various data classification techniques [30]. Due to its suitability in
precaution programming, the model achieved and resulted in great prediction
models. The accuracy of forecasting can vary depending on the approaches used by
various research systems. As a result, identifying equipment capable of providing
extremely accurate diagnosis and tracking for the outbreaks and outbursts are cri-
tical. CI approaches are the most effective policymaking strategies for dealing with
real-world and systemic issues. The goal is to see how different CI techniques
perform when it comes to categorizing COVID-19 samples. The performance of CI
techniques has been evaluated using a variety of categorization performance abil-
ities. The following are five regularly used CI models: artificial neural network
(ANN) for classification, K-nearest neighbor, support vector machine (SVM), lin-
ear regression, and multilayer perceptron.
In the classification of diabetes proteins, machine learning has already been
used as an instance in biological sector [31,32], and in the prediction of diabetes
and heart disease [33]. The ANN, fuzzy logic, K-mean, SVM are some of the best
classifiers used in this reaction [34]. The fuzzy logic classifier approach is an
alternate form of CI methodology that scholars have used to categorize COVID-
19 situations [35]. In [35], the authors propose a classification model for calculating
and projecting the occurrences of COVID-19 outbreak for a period of 10 days
based on previously documented cases in China. The suggested framework is an
enriched adaptive neuro-fuzzy inference system (ANFIS) with flower pollination
algorithm using the Salp swarm algorithm model. The rate of misclassified
instances with contradiction was used to measure the accuracy of their studies, and
by using other performance metrics like determination coefficient (R2), computa-

tional time, MAPE, and RMSRE, and perform reasonably better than the existing
methods. Furthermore, the suggested methodology makes use of two independent
datasets of weekly influenza cases reported in two countries, namely the United
States and China. The outcomes revealed much better performance.
In order to properly predict COVID-19 outbreak, various models have been
developed to extract information from several biomedical datasets by several
researchers. In [36], the authors proposed an ANFIS to evaluate and model time-
series difficulties, and it has had a lot of success in recent applications. It has the
ability to evaluate nonlinearity in time series data and incorporates features from
various algorithms like fuzzy logic, random forest, SVM, and ANN among others.
It has been utilized in a variety of prediction applications. In systematic tasks such
as thinking, identifying, and accomplishing challenging tasks, CI has the ability to
duplicate human intelligence utilizing both computer hardware and software.
Because individual intelligence is multidimensional, CI may accomplish tasks such
as knowledge representation, language understanding, and intellectual acquisition.
To date, the use of CI approaches to detect new behavior and identify future
potential has yielded positive outcomes in a variety of commercial and tourism
situations. Recent current methods, such as machine learning and ANN algorithms,
have demonstrated promising results in the extraction of nonlinear dynamic patterns
from a large dataset. The monitor of rodent reservoirs of zoonotic diseases recently
using CI methods within a disease-related framework has been proved useful and
provided accuracy results [37]. Extended-spectrum-lactamase-producing species [38],
as well as tuberculosis (TB) and gonorrhea outbreaks [39] have been predicted. It is
difficult to predict how people will react to infectious diseases. However, with the
availability of big data and the introduction of CI approaches, dangerous diseases have
been gradually compared to population behavior. Data science and psychosocial
informatics have been in use for the prediction of the pandemic in recent time using
huge data generated from IoT-based systems [40,41]. In this era of COVID-19 out-
break, search engine data was used to correlate infection control epidemics with digital
activity trends (such as Google Trend). To actively monitor the effects of COVID-19
outbreak [42], CI technologies have been used as an analytic tool to improve our
ability to actively track social interactions and accurately forecast the spread.
CI is a cutting-edge technology that helps in combating COVID-19 outbreak
globally and reduces the spread of the pandemic by monitoring the infected
patients. This advancement aids in the improved scanning, monitoring, and fore-
casting of existing and potential patients. The most important aspect of CI is to
identify and diagnose the infection as soon as possible. A CI can be used to make
medicines and vaccines, as well as to relieve medical workers stress. The major two
types of CI algorithms are supervised and unsupervised learning, and the method
incorporating a feature from the training samples through learning. ANN, random
forest, SVM, decision tree, naı̈ve Bayes, ANN, bootstrap aggregating, and
AdaBoost are examples of supervised learning [43].
All these methods could be useful in determining diagnosis accuracy and
prescribing appropriate COVID-19 treatment for infected patients. Predictive
results from these methodologies, in particular, might be utilized to inform autho-

rities and the public ahead of time and offer appropriate models and strategies.
Unsupervised learning methodologies like principal component analysis (PCA)
help scientists to pinpoint some of the primary causes of COVID-19 outbreak
within a geographical area [44]. Unsupervised learning like as K-means may be
used to categorize COVID-19 victims and identify questionable cases, allowing
scientists to focus on these medical conditions. Latent Dirichlet Allocation one of
the leading models can also be used for feature selection from medical data. Deep
learning models have recently been used in medical sectors in areas like prediction,
social network filtering, diagnosis, and computational biology and are thought to be
excellent tools for assessing infectious diseases. CI can quickly examine aberrant
symptoms and other “obvious indications” and inform patients and medical per-
sonnel [45,46]. It allows for faster and more cost-effective policymaking. It tends to
assist in the establishment of new therapy and monitoring of the outbreak instances
through useful strategies. CI aids in the diagnosis of highly contagious cases using
modern medical technologies like automated medical scanners, computed tomo-
graphy (CT), and magnetic resonance imaging.
CI will provide critical information that will help reduce the risk of this out-
break through the use of background knowledge. It could be used to predict pos-
sible outbreak sites, disease propagation, and the need for locations and healthcare
providers during the pandemic. CI is important for the future detection of infections
and illnesses by using previously trained data over data available at numerous
periods. It outlines the characteristics, causes, and conditions that have caused
infection to spread. In the future, this will be a promising strategy for combating
additional epidemics and infectious diseases. It can be used to take preventative
measures and to treat a variety of ailments. CI has the potential to make healthcare
more transparent and responsive in the future.
CI will develop a sophisticated system to track and forecast the outbreak’s
spread automatically [34,35,47]. A genetic algorithm could also be developed to
erase the infection’s esthetic traits. It has the ability to deliver daily alerts to
patients as well as enhanced COVID-19 pandemic follow-up options. CI can
readily detect the virus’s amount of transmission by distinguishing pieces and “hot
areas,” as well as track and monitor the persons’ interactions. It has the ability to
anticipate how the disease will progress in the future and whether it will return. The
social media with other broadcasting channels can be used to monitor and predict
the spread of and pandemic based on existing data, as well as the outbreak’s dan-
gers and the number of infected cases with the nature of fatalities in a particular
location. CI will aid in the identification of the most affected places, citizens, and
communities, as well as the implementation of effective solutions.
CI can be utilized for drug screening based on existing COVID-19 dataset. It is
suitable for the production and development of pharmaceuticals. This discovery
allows for real-time drug screening, whereas normal screening takes a considerable
time, resulting in substantially better results than a person could accomplish [48]. It
can be used to find effective vaccine for the treatment of the outbreak globally, and
a valuable method for classification and prediction of pandemic and speedy
production of vaccines [49]. CI speeds up the manufacture of vaccines and medi-

cines, and it is also useful in clinical trials while vaccines are being developed.
Healthcare workers are overworked as a result of the increase in the spread of
the outbreak and increase in number of infected cases; thus the workload of these
workers can be reduced using CI models [50–52]. This model can be used for
earlier diagnosis, monitoring, prediction, management for better decision-making
and provides level playground for both health professionals and their students
[53,54]. CI has the potential to have a significant impact on clinical outcomes as
well as address additional future issues, easing the burden on clinicians.
2.3 Big data and analytics in battling COVID-19 outbreak
Big data has impacted all areas of people’s lives in the twenty-first century, parti-
cularly the healthcare industries, which include genetics and medicine [19,55]. The
use of electronic health records (EHR) that replace the paper health reports has
resulted in data growth [56]. To take evidence-based actions, big data provides
incentive and insight for specialist, epidemiologists, and physicians, which resulted
in patient safety [57]. In the present era, numerous technologies such as IoT devices
and sensors contribute significantly to the tremendous expansion of big data.
Current developments in computation, storage, and connection of various devices
have resulted in massive datasets. This extracted useful information from the big
data generated from the connected devices are used for the benefits of government,
industry, science, and society.
The adoption of various IoT-based devices and sensors has increased the
number of data collected and yields a huge amount of healthcare [58]. These sen-
sors and devices will be used to generate over 847 zettabytes (ZB) of data by 2021
based on Cisco recent report [59,60]. Conventional methods cannot process, col-
lect, or regulate this kind of data. Hence, the analysis of such data requires BDAs
for huge datasets with various kinds of data [61]. In order to identify hidden pat-
terns, latent relationships, market dynamics, consumer demands, and other critical
business information BDA are very necessary [62]. The ability to scrutinize this
data can assist a company in dealing with significant information that could have a
negative influence on the organization [63].
Hence, BDA’s main purpose is to help industrial organizations improve their
data interpretation and, as a result, make better and more informed judgments.
BDA enables research and specialists to examine such data that is normally not
possible to analyze using traditional approaches [62]. BDA is a technique that
transforms structure and unstructured data into a more intelligible one for investi-
gative purposes. These computational approaches’ algorithms must recognize
shapes, patterns, and relationships in data across several time horizons [64]. The
results of the findings from BDA are presented using tables, three-dimensional
maps, and charts to aid decision-making.
The process of big data implementation becomes difficult in some application
due to the problem of scalability in the operation of some of these systems, and the
data is sophisticated in core operations of these systems [65]. The critical issues in
big data analysis are to obtain meaningful information from huge data that will
necessitate flexible investigative procedures and methods to report timely results.
Since the conventional techniques are not suitable for big data processes compared
with BDA in gain, an insight into the big data is predicted for meaningful results
[66]. As a result, to accommodate parallel data processing, all-encompassing net-
works and modern technologies are required. Furthermore, complex data streams
from various data sources have several arrangements required analytics models to
get better and understandable results from such data [67]. As a result, the existing
processing models are not efficient based on the problem of inefficiency arise
during computing processes [68]. Figure 2.1 depicts the framework for big data–
enabled CI for COVID-19 outbreak.
Although only a few programs can manage big data volumes in a short pro-
cessing period, customers receive input and advice from BDA systems in a long
time [69]. The best of the best systems, on the other hand, cope with huge amounts
of data and data complexity using a complicated trial-and-error process [70]. One
of the big data graphic analytics is the investigative data examination used to
process large datasets collected from an earth structural model [71]. Because the
volume of big data is so large, traditional database management applications cannot
be used to collect and analyze it. Modern warehouse databases, like as Apache
Hadoop, provide the resolution by allowing for the analysis of distributed data.
Epidemic model: projection of

potential effects of the COVID-19
spread
Shared data Disease-oriented S I
repository knowledge
discovery
S I R
Data integration and
preprocessing
AsyI
Image- Digital
based data data
Input Output
OMICS- Wearable
based data sensors
Computational intelligence: information
Sources of COVID-19 related data
and data mining for emergency response
Figure 2.1 COVID-19 pandemic: an integrated big data conceptual model

In the fight against COVID-19 outbreak, big data has played significant roles
and shown to be a powerful tool in this direction [72,73]. Big data in combination
with BDA have shown to be very useful and creating an intriguing possibilities for
fighting this outbreak. The use of these concepts can aid monitoring the spread of
this pandemic and is useful in the areas of treatment, vaccine development, mon-
itoring on infected patient, among others [74]. For example, combining coronavirus
data streams with AI-based models can be used to create simulation models for
epidemic estimation. This will allow health officials to follow the coronavirus’s
progress and better plan for preventive measures [75]. Big data models, with its
ability to aggregate data and use enormous volumes of data for early detection,
further boost the potential COVID-19 disease prediction. Large-scale COVID-
19 studies can also be used with BDA using data from real-world sources like
infected patients to build high-reliability, robust treatment strategies [76,77]. The
healthcare providers will have a better understanding of the virus’ progression
using the BDA for the huge data processing and allow them to respond more
effectively to various diagnoses and therapies.
In the control and monitoring of COVID-19 outbreak, big data plays a critical
role because of its ability to provide a huge amount of data for predicting outbreaks.
Like the case in [78] study, using the big data records from the outbreak in Italy
used to assess the likelihood for outbreaks is critical for the formulation of efficient
disease management efforts. Rather than utilizing a basic and deterministic human-
transmission-based model, the scientists built more sophisticated models that may
appropriately define pandemic dynamics utilizing data provided by the civil
defense database from Italy. The publicly available dataset can serve as another
data source for epidemic prediction to map localities with similar COVID-19 out-
break [79]. The first experiment is a Wuhan inquiry targeted at tracking individual
movement from and to Wuhan, in order for health officials to quarantine the
COVID-19-infected population.
Big data also helps in the prediction of outbreaks on a worldwide scale.
From the perspective of statistical analysis, the outbreak is forecast using
available datasets that, due to a lack of detailed investigations, put the precision
of the fit for accurate prediction under question. Accuracy may be affected by a
variety of factors, including illness cases, population, living situations, and
ecosystems. Inspired by this, researchers in [80] used a large dataset from many
locations and nations to estimate the pandemic like in China and Korea using
logistic model to assess the forecasts’ credibility. Another study [81] used data
generated from South Korea, China, Italy, and Iran by employed Google Trends
to acquire coronavirus-related data. Five environmental regions are used as the
data textual: (1) globally, to ascertain the global interest of the outbreak; (2)
China the sources of the outbreak; (3) South Korea; (4) Italy; and (5) Iran. This
aggregation of data sources will help visualize the pattern of the outbreak and
predict the future outbreak. The outbreak reports from these countries are also
gathered and evaluated using a data optimization model targeted at accurately
predicting everyday cases of coronavirus infection as well as potentially long-
term epidemic forecasts [82].
In [83], datasets collected from confirmed, death, and recovered cases of all
countries from the John Hopkins University repository were used for prediction
model using ML algorithm. The research was for short-term intervals of 2 weeks in
India. The proposed data analytic can be extended to create larger model for long-
term prediction to estimate the outbreak theoretically. Meanwhile, the American
large-scale datasets were used for the purpose of prediction in the United States
using data analytical approach for a proper investigation of the outbreak [84]. The
main aim of the analysis is to improve the accuracy of the future estimation for any
likely pandemic-like coronaviruses, and fitting the prediction errors in order to
refine the model.
Big data has been used in the monitoring of COVID-19 spread in order to help
the healthcare sectors and government to successfully manage the spread of the
pandemic [75,85]. This has also been used to facilitate the tracking of the COVID-
19 distribution with a variety of newly emerging solutions. For example, the study
in [86] proposed the use of BDA model to track the pandemic spread from the huge
generated data from the outbreak. The China National Health Commission pro-
vided a large dataset of 854,424 flight travelers that flew from Wuhan Airport with
55 airlines to various parts of the 49 cities in China between December 2019 and
January 2020. Using predicted factors such as the local inhabitants and air travelers
to measure the variance of recorded cases in China cities, a multiple linear model is
constructed. More precisely, the authors used a Spearman correlation analysis with
the number of 49 reported cases from Wuhan using travelers’ traffic for a period of
time. The results of the finding have shown a high correlation between the popu-
lation and the infection cases within the cities in China. In [87], the Geographic
Information Systems (GIS) was used with big data for spatial analysis to promote
heterogeneous data integration from healthcare data tools like the data from the
patients, patients, clinical laboratories, and the public.
Jin et al. [88] provided a detailed guide with valuable methods for the treatment
and diagnosis of COVID-19. The techniques of the guiding principle, biological
aspects, disease prevention, diagnosis, and COVID-19 disease care are all included in
this document. From the 11,500 people tested by Zhongnan Hospital of Wuhan
University as a pilot trial, 276 were discovered as potentially infected casualties, with
170 being diagnosed. A huge set of dataset has been investigated using clinical tests,
including CT/X-ray imaging, CT/X-ray imaging in response to pathogen detection in
the respiratory tract, and manifestations to hematological analysis.
The development of vaccine is very paramount in the prevention of individuals
from the burden of infectious diseases and pandemic. In the identification of better
vaccine for the treatment of COVID-19 outbreak, big data plays prominent roles.
The use of this huge data within this short time has really helped in the production
of suitable vaccine for COVID-19 outbreak globally. The use of GISAID database
in the work of [89] has aided in the removing residues amino acids. The goal of
research is to find effective opportunities for the growth of vaccines and medica-
tions to combat the COVID-19 pandemic. Another vaccine research initiative by
[90] compares the spike proteins in SARS-CoV, MERS-CoV, and SARS-CoV-2
with other coronavirus strains in four humans previously isolated. This research
will aid in the essential screening of the SARS-CoV-2 spike sequence and structure,
which could aid in the development of an effective vaccine.
In the production of drugs manufacturing models to battle COVID-19 pan-
demic, big data has played prominent roles. For instance, [91] proposed a mole-
cular docking-based solution for drug research in fighting COVID-19 outbreak.
The first screening involved more than 2,500 small molecules and was validated
through the use of Glide molecular docking software in the FDA-approved drug
database. In the experiment, 15 of the 25 verified medicines exhibited considerable
inhibitory properties. The drug formulation toward COVID-19 is triggered by the
inhibitory incremental change and chronic inflammation. In addition, the authors in
[92] used big data to develop a drug repositioning strategy to combat the COVID-
19 epidemic. For the development of COVID-19 vaccines, the authors use ML to
incorporate both the information graph and the literature.
In addition, reverse vaccinology and immunological computer science can be
employed to produce vaccines for the global COVID-19 outbreak [93]. The
National Center for Biotechnology Information’s online database selected SARS-
CoV-2 strain to analyze various entries while Immune Epitope Database online
epitope was utilized for B-cell and T-cell prediction. To classify the possible vac-
cine constructions from the produced dataset, a number of procedures may be
required, including immunogenicity, genotoxicity, and biophysical features of
products. The National Center for Biotechnology Knowledge for Promoting the
Manufacture of Vaccines [94] has also acquired a large dataset. Different peptides
have been suggested for the two-step production of a new COVID-19 vaccine. To
begin, the entire genome of COVID-19 was examined using a comparative geno-
mic method to identify the most likely antigenic target.
2.4 The limitations of using big data and computational

intelligence to fight the COVID-19 pandemic
COVID-19 and other infectious illnesses have the potential to benefit from CI. In
contrast, the benefits of AI interventions will take time to manifest because AI
systems are still in their infancy [95]. In its study of the use of CI toward COVID-
19, the authors in [96] stated that just a few of the CI programs studied are oper-
ationally mature at this moment. For efficient prediction, CI needs a large amount
of data; however, in a circumstance when the data is scarce, the model’s overall
performance will be limited. To train CI and AI models, there are (yet) insufficient
public data and open datasets relevant to the COVID-19 pandemic. Other potential
obstacles include non-adjustment procedures, big data arrogance, a torrent of sci-
entific results, and external data that must be swapped for a better CI model.
After the outbreak is gone, the key concerns are issues of data protection for the
general public to ensure that the data collected is not used for other purposes by
policy-makers, and the degradation is not pushed back in the fight against the COVID-
19 pandemic. Though coronavirus diseases could be practically eradicated, as pre-
dicted by the authors in [97], some data-hungry governments may claim that biometric
tracking techniques must be kept in place in order to prepare for another pandemic
outbreak, or second wave of COVID-19-related diseases, as we have seen in India and
other Asian nations. An instance of another outbreak of Ebola in the central region of
Africa requires the emerging of using modern tracking mechanisms to be put in place
for any emergency. CI-based radiological innovation has been considerably adopted
by several Chinese clinics, and the promise of CI-based models’ complete adoption
magically reappears. Radiologists have highlighted concerns about a dearth of data for
training CI-based models, and the bulk of the outbreak samples accessible from
country medical institutes does not include sample preference. The use of CT scans
and X-rays has the tendency to escalate infection and contaminate the instruments.
The use of crowded social platforms, big data hubris, and outlier data are still
not reliable and valid due to a lack of evidence, and technological challenges in
COVID-19 outbreak forecasting [98]. CI techniques are not used in the bulk of
monitoring and forecasting frameworks currently in use. Rather, most analysts
consider SIR models, which are epidemiologically validated models. The use of
immunological SIR model for instance has been adopted by Robert Koch Institute
in Berlin that includes government controls like curfews, sterilization, and social
disassociating advice. This approach was created in China to demonstrate that
isolation might lower the rate of incidence below the catastrophic value [99].
Table 2.1 summarizes the CI issues in the COVID-19 epidemic.
Table 2.1 Computational intelligence limitations
S. Issues in the use of CI Explanations

no.
1 It necessitates competent It is devoid of any further details, such as a person, to
monitoring depict the exact outcome. It necessitates meticulous
data collection and monitoring; alternatively, profes-
sionals and physicians will be unable to produce
accurate results
2 Not allowing for creative Only exact data will provide adequate enactment. Human
thought critical judgment is not taken into consideration
employing the model, which cannot fit the COVID-19
definition of innovative thinking as a human being
does. This model does not give room for latest
decisions to be made in the absence of data
3 It only learns from the facts CI always works with the information provided by the
it is given patient. The information gathered is used to create a
precise COVID-19 diagnosis and forecast
4 Important for the AI The approximation used by CI model is a drawback in the
algorithm’s treatment accuracy percentage, a critical measure for proper care
and more so. Predictions could include whether a CI’s
function is absolute or if more data and training are
necessary over time
5 Does not comprehend the CI methods to decision-making do not take into considera-
human feeling tion human emotions, reasoning, or motivations. CI’s
models do exactly what they are intended to do
The present pandemic has brought serious challenges because it can spread
quickly with human interaction and among people traveling internationally; thus
there is total ban on inter-traveling among several nations [100,101]. Due to these,
foreign visitors are forced to supply their full details like personal information,
travel history, reason for their visiting, and place of residence, and most impor-
tantly there is imposed quarantine restriction globally [102]. The Chinese media
secretly published visitors secret information is an example of such cases; this
explains that the importance of visitor information can cause dispute-of-law in any
nation. (1) The important and distinct definitions of right to personal data have been
clearly defined by the EU, United States, and China; (2) the growth to centralized
approach to relevant laws by these three nations points to the fact that healthcare
privacy law is to be a contractual law; and (3) the de-Americanization is aggres-
sively advocated for and has important data privacy laws. This added to how
international information regulation is being applied or the patterns and methods of
administration of the privacy data laws [1,70].
In the making of CI-based model and BDA, the lack of standard datasets poses
a significant challenge for their use in combating the COVID-19 pandemic.
Various CI models using huge datasets have been discussed in this literature, but
none have been assessed using the same dataset. Thus, to draw conclusion becomes
difficult due to the use of two datasets with different quantities of samples for
COVID-19 outbreak prediction and detection. In comparison, the majority of
datasets in the literature were created as a result of individual efforts, such as when
writers acquire information on the Internet then combine them to form their own
dataset and test the methods they suggest. Healthcare organization like WHO and
CDC plays important roles in resolving this issue as well as the governments, and
large since they have accessed to high-quality and massive datasets. Medical data
like X-ray and CT scans can be obtained from hospitals, reports from self-diagnosis
applications, satellite data, and personal information.
The CORD-19 dataset is being managed by various collaborators with the original
owner Georgetown Security Center like Microsoft Research, Chan Zuckerberg
Initiative, Allen Institute for AI, and National Institutes of Health [103]. Also, in China,
Alibaba DAMO Academy has partnered with various hospitals to create AI-based
models for COVID-19 case prediction and identification, and more than 5,000 instan-
ces of COVID-19 have been reported, and hospitals are responsible for delivering CT
scans. As recorded in [104] thanks to its remarkable performance, this device has been
used by more than 20 hospitals in China: 96% accuracy in just 20 s. Data scientists
encounter numerous challenges when working with big data. One challenge is com-
piling, integrating, and storing big data derived from various sources [64,65]. Another
vital problem is the administration and application of this generated data in healthcare
systems. To maximize the effectiveness of big data in utilization in medical case
required the use of a reliable BDA like CI model to facilitate the perception of a
dependable data. BDA is built on the foundation of good data management.
The utilization of huge data generated from the use of various techniques like
sensors, devices, and patients’ symptoms using healthcare internal infrastructure is
another difficulty in generating a reliable result from huge data [16]. The majority
of the time, analyzing data generated within businesses, is insufficient. We need to

take it a step further by linking both internal and external data together to get relevant
information and knowledge. Third-party sources, market swings, weather forecasts
and traffic problems, social network data, client feedback, and resident input are all
examples of external data. The principle of participatory and incremental learning is
being applied in big data and analytics. They address a range of difficulties that arise
during data learning like resource constraints and data accessibility challenges, thus,
resulted in obtaining useful information for a better decision-making in healthcare
system. When iterative learning is applied to data, it leads to a faster way of pre-
dicting new data streams. When the drift notion is unavailable, the incremental
algorithm is strongly suggested. On the other hand, using ensemble techniques to
produce reliable results with a big drift notion is highly suggested.
Another key issue is analogical thinking, which is intrinsically unreliable when
it comes to disease predictions. In big data and CI, there is always an underlying
assumption that their applications in various fields create complicated simulation
models that sidestep the need to acquire fundamental epidemiological details.
Nonetheless, a trustworthy fact on the scenario counts and the biotic processes that
underlie a pandemic for evolving COVID-19 outbreaks, let alone the behavioral
reactions of infected individuals, is still lacking, making it harder to alter or com-
prehend precise complicated prototypes on spatiotemporal scales relevant to
decision-making. Due to the requirement for versatile modules that enable quick
reactions in light of the high degree of controversy surrounding emergency situa-
tions, it is hoped that the most effective systems would stay basic [105]. According
to epidemiological studies, simple models are easier to be understood, expressed,
and interpreted than complex one, especially in healthcare system [106].
A lack of a clear message risks two undesirable outcomes since most politi-
cians lack in-depth modeling experience. Thinking that without skepticism and
options, the technique would be misguided, or ignoring and there is no proof that
we must control epidemics as effectively and feasible. During epidemics, choices
must be taken rapidly and piecemeal, and erroneous data models may be effective
tools for doing so. In light of the aforementioned challenges, computational power
of CI model can be utilized to steadily improve the power of big data in healthcare
system. These will provide genuine bravery for improved monitoring and bring
about some beneficial prediction systems [107,108].
The use of BDA and CI models with information-sharing platforms is being
used to protect personal data security and privacy [109]. Rapid data transmission
and communication among geographically distributed responder teams are enabled
via an Internet connection. Methodologically, effective modeling solutions that
incorporate diverse predictions to reduce uncertainty are being created [110]. An
innovative, collaborative approach has emerged in response to the outbreak of
COVID-19 across academic organizations: for example, using Twitter and other
channels to communicate, assess, and openly debate the consequences of new
research as it emerges. Ironically, during a crisis, controlling the misinformation
that is already circulating on social media is likely to become one of the most
pressing problems for restricting the epidemic.
These technologies will remain dislocated and unworkable unless the aforemen-
tioned issues are rectified. It appears that all three problems may be aided by directing
the majority of funds and knowledge to the communities most vulnerable to epi-
demics. Methodologically, effective modeling solutions that incorporate diverse pre-
dictions to reduce uncertainty are being created. Using modern big data technologies,
nimble, distributed groups of people will be required to cover the systematic and
functional components of the epidemic response. To complement and explain the
restricted availability of socio-geographical data needed to track the progress of sig-
nificant forecasts. The methods can be used to forecast the current pandemic so that
the policy-makers can plan ahead of another similar outbreak or epidemic of this kind.
The combination of both BDA and CI models can be troublesome if the available data
for modeling is very small since CI model depends meanly on a huge amount of data
to be able to diagnosed, predicted, and classifier efficiently and properly.
The fulfillment of these methods in healthcare systems significantly depends
on serious investment on training, development of relevant models that can be used
to tackle any infectious diseases. Furthermore, there should constant release of fund
to fill the critical gap in analytics process, especially in the developing nation. The
social media sites, X-ray images, scanning images, phones recordings among others
can be a source of generating big data in medical systems with public available data
[111], and their text or video can be used in a variety of ways. In the context of
COVID-19 outbreak, big data can inform of generated information about the epi-
demic locations, medical records, infected case history, scanned and X-ray images
among others. Big data and analytics have been proving useful in fighting against
COVID-19 pandemic [73].
2.5 The practical case of using computational

intelligence in fighting COVID-19 pandemic
Data collection and synthesis on the CXR images of COVID-19 dataset are used for
the study. The conversion of RGB image into grayscale images was done using
image-processing techniques. The feature textual descriptor was performed using a
histogram of oriented gradient (HOG) [112,113], and PCA was used for feature
selection [114]. The suggested methodology for identifying and classifying
COVID-19 includes a train test split of the data, as well as performance evaluation.
The proposed model was used for the classification of frontal CXR pictures into
COVID-19 infected, and pulmonary tuberculosis (PTB) disorders. Figure 2.2
depicts a four-phased approach.
The dataset from frontal CXR image was used in the chapter for the purpose of
classifiers of the dataset into COVID-19 and PTB illness and was obtained from
[115] with the sample sizes of 527, 406, and 394 for COVID-19, NORMAL, and
PTB, respectively. The feature extraction approach is used to remove linguistic
information from the dataset because it is an image (HOG). These lungs were
employed in the data categorization experiment. There are presently 1,327 frontal
CXR pictures in the collection, separated into three classification classes. The test
75:25
CXR images HOG PCA train test Extra trees
or COVID-19, split model
normal
and PTB
Figure 2.2 The workflow of the framework
Confusion matrix
120
COVID19 129 4 1
100
80
True label
Normal 23 65 14
60
40
PTB 22 5 69 20
COVID19 Normal PTB

predicted label
Figure 2.3 Confusion matrix for ET model
result was based on the split set, which represents 25% of the entire dataset because
the dataset was split. This experiment was done out on a Windows computer
operating systems running on an Intel Core i3-4200 CPU at 2.50–2.70 GHz and 4
GB RAM, with R-programming language for executing the programming code.
The Extra Trees algorithm yielded a result of 79.22% accuracy, 81.00% precision,
79.00% recall, and 79.02% F1-score. The best features from the generated HOG
features were chosen using the PCA technique, resulting in a better outcome.
2.5.1 Confusion matrix

The performance of each class is depicted in the confusion matrix in Figure 2.3.
Out of 134 cases of COVID-19 disease, 129 (96%) were accurately recognized as
COVID-19 disease, 4 (3%) as NORMAL, and 1 (1%) as PTB disease. In the case of
NORMAL lungs, 65 (64%) of the 102 cases were accurately classified as
NORMAL, 23 (23%) as COVID-19, and 14 (14%) as PTB. Furthermore, 22 (23%)
of PTB patients were improperly classified as COVID-19, 5 (5%) as NORMAL
lungs, and 69 (72%) of 96 cases were incorrectly classified as NORMAL lungs but
were accurately diagnosed as PTB.
2.5.2 ROC curves

As demonstrated in Figure 2.4, ROC curves reflect the trade-off between TPR and FPR.
It is utilized to see how well each class does in terms of recall and training. The top
leftmost corner of the ROC architecture corresponds to an ideal data classification model
that returns 100% TPR and 0% FPR. As a result, a solid data categorization should
produce a point in the ROC layout’s upper left corner [113]. The ROC values are all
close to one, suggesting exceptional classification performance. The COVID-19 ROC
value for Class is 95.02%, NORMAL is 92.24%, and PTB is 90%.
2.5.3 Precision-recall curve

As shown in Figure 2.5, this indication reflects the trade-off between precision and
recalls for each class. In most cases, high precision comes at the expense of low
ROC curves
1.0
0.8
True positive rate
0.6
0.4
ROC curve of class COVID-19 (area = 0.97)
ROC curve of class NORMAL (area = 0.99)
0.2 ROC curve of class PULMONARY (area = 0.98)
Micro-average ROC curve (area = 0.98)
Micro-average ROC curve (area = 0.98)
0.0
0.0 0.2 0.4 0.6 0.8 1.0
False positive rate
Figure 2.4 ROC curve for ET model
Precision-recall curve
1.0
0.8
Precision
0.6
0.4
Precision-recall curve of class COVID-19 (area = 0.936)
0.2 Precision-recall curve of class NORMAL (area = 0.846)
Precision-recall curve of class PTB (area = 0.787)
Micro-average precision-recall curve (area=0.846)
0.0
0.0 0.2 0.4 0.6 0.8 1.0
Recall
Figure 2.5 ET model precision-recall curve (PRC)

recall, or vice versa. A good data categorization model should be able to keep both
precision and recall high [113]. Class has a COVID-19 PRC value of 0.936,
NORMAL of 0.846, and PTB of 0.787.
2.6 Conclusion
COVID-19 is a one-of-a-kind coronavirus from the pneumonia family. It is vital to
have a reliable method for detecting COVID-19 in infected patients, and CI has
proven to be effective. This chapter looked at various recent COVID-19 control
strategies that used CI. The majority of the studies we looked at focused on
COVID-19 pandemic diagnosis, prediction, and monitoring, as well as image
classification with CXR images and CTs for COVID-19 outbreak. While the bulk
of the recent studies employed either machine learning/deep learning to extract and
choose features, a handful of them combined the two approaches by using ML to
extract and select features and DL for classification and prediction. There has been
effort to improve the accuracy of CI algorithms, there are still CI-based diagnostic
methods that yield positive results, but still there is no readymade answer in this
direction. The system developed using CI models has been proved useful in mon-
itoring, diagnosis, and prediction of COVID-19 outbreak in developed nations. The
model works on the provided data that is related to COVID-19 signs and symptoms.
The models that this has improved have seriously helped physicians in dealing with
infected patients and the control of the spread of COVID-19 pandemic globally.
This helps in early diagnosis of COVID-19 patients, vaccine development, and the
monitoring of isolated patients. Therefore, the use of BDA and CI algorithms has
really help in the diagnosis, prediction, classification, and treatment during this
outbreak. Though the use of BDA and CI models has not be fully integrated to
really combat infectious diseases, its use can never be overemphasized and played
substantial role in the development of vaccines for the treatment of any infectious
disease. The findings of this chapter show that BDA and CI algorithms place pro-
minent roles in fighting COVID-19 pandemic globally, and their direct application
can reduce the spread of this pandemic within nations.
References
[1] Lukman, A. F., Benedicta, A., Awotunde, J. B., et al. (2022). Robust
statistical modeling of COVID-19 prevalence in African epicentres.
Studies in Systems, Decision and Control, 366, 315–358.
[2] Dunford, D., Dale, B., Stylianou, N. L., Ahmed, M. and Arenas, IdlT,
(2020). Coronavirus: the world in lockdown in maps and charts. BBC News
(available at https://www.bbc.com/news/world-52103747).
[3] Awotunde, J. B., Jimoh, R. G., AbdulRaheem, M., Oladipo, I. D., Folorunso,
S. O. and Ajamu, G. J. (2022). IoT-based wearable body sensor network for
COVID-19 pandemic. In Advances in Data Science and Intelligent Data
Communication Technologies for COVID-19 (pp. 253–275). Cham: Springer.
[4] Guan, W. J., Ni, Z. Y., Hu, Y., et al. (2020). Clinical characteristics of
coronavirus disease in 2019 in China. New England Journal of Medicine,
382(18), 1708–1720.
[5] Roy, S. and Chakraborty, C. (2021). Panic buying situation during COVID-
19 global pandemic. Journal of Information Technology Management, 13(2),
231–244.
[6] Garg, L., Chakraborty, C., Mahmoudi, S. and Sohmen, V. S. (2021).
Healthcare informatics for fighting COVID-19 and future pandemics. In
EAI/Springer Innovations in Communication and Computing. Springer
Nature. ISBN: 978-3-030-72751-2.
[7] Shelke, Y. and Chakraborty, C. (2020). Augmented reality and virtual reality
transforming spinal imaging landscape: a feasibility study. IEEE Computer
Graphics and Applications, 41(3), 124–138.
[8] Jang, S., Lee, S., Choi, S. M., Seo, J., Choi, H. and Yoon, T. (2016).
Comparison between SARS CoV and MERS CoV using apriori algorithm,
decision tree, SVM. In MATEC Web of Conferences (Vol. 49, p. 08001).
EDP Sciences.
[9] Awotunde, J. B., Folorunso, S. O., Bhoi, A. K., Adebayo, P. O. and Ijaz, M. F.
(2021). Disease diagnosis system for IoT-Based wearable body sensors with
machine learning algorithm. In Hybrid Artificial Intelligence and IoT in
Healthcare (pp. 201–222). Singapore: Springer.
[10] Brown, D. E., Abbasi, A. and Lau, R. Y. (2015). Predictive analytics:
predictive modeling at the micro level. IEEE Intelligent Systems, 30(3),
6–8.
[11] Chakraborty, C. and Abougreen, A. N. (2021). Intelligent internet of things
and advanced machine learning techniques for COVID-19. EAI Endorsed
Transactions on Pervasive Health and Technology, 7(26), e1.
[12] Ahmed, M. B., Boudhir, A. A., Santos, D., El Aroussi, M. and Karas, İ. R.
(Eds.). (2020). Innovations in smart cities application. In The Proceedings of
the 4th International Conference on Smart City Applications (3rd ed.).
Springer Nature.
[13] Rodgers, W. (2020). Artificial Intelligence in a Throughput Model: Some
Major Algorithms. CRC Press.
[14] Priyanka, K. and Kulennavar, N. (2014). A survey on big data analytics in
health care. International Journal of Computer Science and Information
Technologies, 5(4), 5865–5868.
[15] Kruse, C. S., Goswamy, R., Raval, Y. J. and Marawi, S. (2016). Challenges
and opportunities of big data in health care: a systematic review. JMIR
Medical Informatics, 4(4), e38.
[16] Oussous, A., Benjelloun, F. Z., Lahcen, A. A. and Belfkih, S. (2018). Big
data technologies: a survey. Journal of King Saud University-Computer and
Information Sciences, 30(4), 431–448.
[17] Olaronke, I. and Oluwaseun, O. (2016, December). Big data in healthcare:
prospects, challenges and resolutions. In 2016 Future Technologies
Conference (FTC) (pp. 1152–1157). IEEE.
[18] Dai, H. N., Wang, H., Xu, G., Wan, J. and Imran, M. (2020). Big data ana-
lytics for manufacturing Internet of Things: opportunities, challenges and
enabling technologies. Enterprise Information Systems, 14(9–10), 1279–1303.
[19] Hulsen, T., Jamuar, S. S., Moody, A., et al. (2019). From big data to preci-
sion medicine. Frontiers in Medicine, 6, 34.
[20] Austin, C. and Kusumoto, F. (2016). The application of Big Data in medi-
cine: current implications and future directions. Journal of Interventional
Cardiac Electrophysiology, 47(1), 51–59.
[21] Galetsi, P., Katsaliaki, K. and Kumar, S. (2019). Values, challenges and
future directions of big data analytics in healthcare: a systematic review.
Social Science & Medicine, 241, 112533.
[22] Lacroix, P. (2019). Big data privacy and ethical challenges. In Big Data, Big
Challenges: A Healthcare Perspective (pp. 101–111). Springer, Cham.
[23] Hu, Z., Ge, Q., Jin, L. and Xiong, M. (2020). Artificial intelligence fore-
casting of COVID-19 in China. arXiv preprint arXiv:2002.07112.
[24] Bai, H. X., Hsieh, B., Xiong, Z., et al. (2020). Performance of radiologists in
differentiating COVID-19 from viral pneumonia on chest CT. Radiology.
[25] Zurada, J. M., Marks, R. J. and Robinson, C. J. (1994). Computational
Intelligence Imitating Life (No. CONF-940699). Institute of Electrical and
Electronics Engineers, Inc., New York, NY.
[26] Siddique, N. and Adeli, H. (2013). Computational Intelligence: Synergies of
Fuzzy Logic, Neural Networks and Evolutionary Computing. John Wiley &
Sons.
[27] Awotunde, J. B., Matiluko, O. E. and Fatai, O. W. (2014). Medical diag-
nosis system using fuzzy logic. African Journal of Computing & ICT, 7(2),
99–106.
[28] Ayo, F. E., Folorunso, S. O., Abayomi-Alli, A. A., Adekunle, A. O. and
Awotunde, J. B. (2020). Network intrusion detection based on deep learning
model optimized with rule-based hybrid feature selection. Information
Security Journal, 29(6), 267–283.
[29] Ayo, F. E., Ogundokun, R. O., Awotunde, J. B., Adebiyi, M. O. and Adeniyi,
A. E. (2020). Severe acne skin disease: a fuzzy-based method for diagnosis.
In Lecture Notes in Computer Science (including subseries Lecture Notes in
Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 12254
LNCS, pp. 320–334). Cham: Springer.
[30] Oladele, T. O., Ogundokun, R. O., Awotunde, J. B., Adebiyi, M. O. and
Adeniyi, J. K. (2020). Severe acne skin disease: a fuzzy-based method for
diagnosis. In Lecture Notes in Computer Science (including subseries
Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
(Vol. 12254 LNCS, pp. 428–441). Cham: Springer.
[31] Kurd, Z., Kelly, T. and Austin, J. (2007). Developing artificial neural net-
works for safety-critical systems. Neural Computing and Applications, 16(1),
11–19.
[32] Berglund, E. and Sitte, J. (2006). The parameterless self-organizing map
algorithm. IEEE Transactions on Neural Networks, 17(2), 305–316.
[33] Oladipo, I. D., Babatunde, A. O., Awotunde, J. B. and Abdulraheem, M.

(2021). An improved hybridization in the diagnosis of diabetes mellitus
using selected computational intelligence. In Communications in Computer
and Information Science (Vol. 1350, pp. 272–285).
[34] Awotunde, J. B., Ayo, F. E., Jimoh, R. G., et al. (2020). Prediction and clas-
sification of diabetes mellitus using genomic data. In Intelligent IoT Systems in
Personalized Health Care (pp. 235–292). London: Academic Press.
[35] Al-Qaness, M. A., Ewees, A. A., Fan, H. and Abd El Aziz, M. (2020).
Optimization method for forecasting confirmed cases of COVID-19 in
China. Journal of Clinical Medicine, 9(3), 674.
[36] Jang, J. S. (1993). ANFIS: adaptive-network-based fuzzy inference system.
IEEE Transactions on Systems, Man, and Cybernetics, 23(3), 665–685.
[37] Han, B. A., Schmidt, J. P., Bowden, S. E. and Drake, J. M. (2015). Rodent
reservoirs of future zoonotic diseases. Proceedings of the National Academy
of Sciences of the United States of America, 112(22), 7039–7044.
[38] Goodman, K. E., Lessler, J., Cosgrove, S. E., et al. (2016). A clinical deci-
sion tree to predict whether a bacteremic patient is infected with an
extended-spectrum b-lactamase-producing organism. Clinical Infectious
Diseases, 63(7), 896–903.
[39] Wong, Z. S., Zhou, J. and Zhang, Q. (2019). Artificial intelligence for infec-
tious disease big data analytics. Infection, Disease & Health, 24(1), 44–48.
[40] Mahroum, N., Adawi, M., Sharif, K., et al. (2018). Public reaction to
Chikungunya outbreaks in Italy—insights from an extensive novel data streams-
based structural equation modeling analysis. PLoS One, 13(5), e0197337.
[41] Bragazzi, N. L., Alicino, C., Trucchi, C., et al. (2017). Global reaction to the
recent outbreaks of Zika virus: insights from a big data analysis. PLoS One,
12(9), e0185263.
[42] Sujata, D., Chinmay, C., Sourav, K. G. and Subhendu, K. P. (2021).
Intelligent computing on time-series data analysis and prediction of COVID-
19 pandemics, Pattern Recognition Letters, 151, 69–75, https://doi.org/10.
1016/j.patrec.2021.07.027.
[43] Folorunso, S. O., Awotunde, J. B., Ayo, F. E., and Abdullah, K. K. A.
(2021). RADIoT: the unifying framework for IoT, radiomics and deep
learning modeling. In Hybrid Artificial Intelligence and IoT in Healthcare
(p. 109–128). Singapore: Springer.
[44] Wang, Y., Fan, Y., Bhatt, P. and Davatzikos, C. (2010). High-dimensional
pattern regression using machine learning: from medical images to con-
tinuous clinical variables. NeuroImage, 50(4), 1519–1535.
[45] Ai, T., Yang, Z., Hou, H., et al. (2020). Correlation of chest CT and RT-PCR
testing for coronavirus disease 2019 (COVID-19) in China: a report of 1014
cases. Radiology, 296, E32–E40.
[46] Luo, H., Tang, Q. L., Shang, Y. X., et al. (2020). Can Chinese medicine be
used for prevention of corona virus disease 2019 (COVID-19)? A review of
historical classics, research evidence and current prevention programs.
Chinese Journal of Integrative Medicine, 1–8.
[47] Muhammad, L. J., Algehyne, E. A., Usman, S. S., Ahmad, A., Chakraborty,
C. and Mohammed, I. A. (2021). Supervised machine learning models for
prediction of COVID-19 infection using epidemiology dataset. SN Computer
Science, 2(1), 1–13.
[48] Haleem, A., Vaishya, R., Javaid, M. and Khan, I. H. (2020). Artificial
intelligence (AI) applications in orthopaedics: an innovative technology to
embrace. Journal of Clinical Orthopaedics & Trauma, 11, S80–S81.
[49] Bobdey, S. and Ray, S. (2020). Going viral – Covid-19 impact assessment: a
perspective beyond clinical practice. Journal of Marine Medical Society,
22(1), 9.
[50] Gozes, O., Frid-Adar, M., Greenspan, H., et al. (2020). Rapid AI develop-
ment cycle for the coronavirus (COVID-19) pandemic: initial results for
automated detection & patient monitoring using deep learning CT image
analysis. arXiv preprint arXiv:2003.05037.
[51] Pirouz, B., Shaffiee Haghshenas, S., Shaffiee Haghshenas, S. and Piro, P.
(2020). Investigating a serious challenge in the sustainable development
process: analysis of confirmed cases of COVID-19 (new type of cor-
onavirus) through a binary classification using artificial intelligence and
regression analysis. Sustainability, 12(6), 2427.
[52] Smeulders, A. W. and Van Ginneken, A. M. (1989). An analysis of pathol-
ogy knowledge and decision making for the development of artificial
intelligence-based consulting systems. Analytical and Quantitative Cytology
and Histology, 11(3), 154–165.
[53] Hussain, A. and do Vale Moreira, N. C. (2020). Clinical considerations for
patients with diabetes in times of COVID-19 pandemic. Diabetes &
Metabolic Syndrome, 14(4), 451.
[54] Gupta, R., Ghosh, A., Singh, A. K. and Misra, A. (2020). Clinical con-
siderations for patients with diabetes in times of COVID-19 epidemic.
Diabetes & Metabolic Syndrome, 14(3), 211.
[55] Riba, M., Sala, C., Toniolo, D. and Tonon, G. (2019). Big data in medicine,
the present and hopefully the future. Frontiers in Medicine, 6, 263.
[56] Dinh-Le, C., Chuang, R., Chokshi, S. and Mann, D. (2019). Wearable health
technology and electronic health record integration: scoping review and
future directions. JMIR mHealth and uHealth, 7(9), e12861.
[57] McCue, M. E. and McCoy, A. M. (2017). The scope of big data in one
medicine: unprecedented opportunities and challenges. Frontiers in
Veterinary Science, 4, 194.
[58] Tariq, N., Asim, M., Al-Obeidat, F., et al. (2019). The security of big data in fog-
enabled IoT applications including blockchain: a survey. Sensors, 19(8), 1788.
[59] Adeniyi, E. A., Ogundokun, R. O. and Awotunde, J. B. (2021). IoMT-based
wearable body sensors network healthcare monitoring system. In IoT in
Healthcare and Ambient Assisted Living (pp. 103–121). Springer, Singapore.
[60] Hofdijk, J., Séroussi, B., Lovis, C., Sieverink, F., Ehrler, F. and Ugon, A.
(2016). Transforming healthcare with the Internet of Things. In Proceedings
of the EFMI Special Topic Conference.
[61] Mital, R., Coughlin, J. and Canaday, M. (2015). Using big data technologies
and analytics to predict sensor anomalies. In AMOS (p. 84).
[62] Berman, E., Felter, J. H. and Shapiro, J. N. (2020). Small Wars, Big Data:
The Information Revolution in Modern Conflict. Princeton University Press.
[63] Raghupathi, W. and Raghupathi, V. (2014). Big data analytics in healthcare:
promise and potential. Health Information Science and Systems, 2(1), 3.
[64] Teijeiro, D., Pardo, X. C., González, P., Banga, J. R. and Doallo, R. (2018).
Towards cloud-based parallel metaheuristics: a case study in computational
biology with differential evolution and spark. The International Journal of
High Performance Computing Applications, 32(5), 693–705.
[65] Candela, L., Castelli, D. and Pagano, P. (2012). Managing big data through
hybrid data infrastructures. ERCIM News, 89, 37–38.
[66] Marjani, M., Nasaruddin, F., Gani, A., et al. (2017). Big IoT data analytics:
architecture, opportunities, and open research challenges. IEEE Access, 5,
5247–5261.
[67] Assuncao, M. D., Calheiros, R. N., Bianchi, S., Netto, M. A. and Buyya, R.
(2013). Big data computing and clouds: challenges, solutions, and future
directions. arXiv preprint arXiv:1312.4722, 10.
[68] Hashem, I. A. T., Yaqoob, I., Anuar, N. B., Mokhtar, S., Gani, A. and Khan,
S. U. (2015). The rise of “big data” on cloud computing: review and open
research issues. Information Systems, 47, 98–115.
[69] Singh, D. and Reddy, C. K. (2015). A survey on platforms for big data
analytics. Journal of Big Data, 2(1), 8.
[70] Awotunde, J. B., Folorunso, S. O., Jimoh, R. G., Adeniyi, E. A., Abiodun, K.
M. and Ajamu, G. J. (2021). Application of artificial intelligence for
COVID-19 epidemic: an exploratory study, opportunities, challenges, and
future prospects. In Artificial Intelligence for COVID-19 (pp. 47–61).
Springer, Cham.
[71] Steed, C. A., Ricciuto, D. M., Shipman, G., et al. (2013). Big data visual
analytics for exploratory earth system simulation analysis. Computers &
Geosciences, 61, 71–82.
[72] Chae, S., Kwon, S. and Lee, D. (2018). Predicting infectious disease using
deep learning and big data. International Journal of Environmental Research
and Public Health, 15(8), 1596.
[73] Bansal, S., Chowell, G., Simonsen, L., Vespignani, A. and Viboud, C.
(2016). Big data for infectious disease surveillance and modeling. The
Journal of Infectious Diseases, 214(Suppl. 4), S375–S379.
[74] Eisenstein, M. (2018). Infection forecasts powered by big data. Nature,
555(7695).
[75] Buckee, C. (2020). Improving epidemic surveillance and response: big data
is dead, long live big data. The Lancet Digital Health, 2(5), e218–e220.
[76] Awotunde, J. B., Jimoh, R. G., Oladipo, I. D., Abdulraheem, M., Jimoh, T.
B. and Ajamu, G. J. (2021). Big data and data analytics for an enhanced
COVID-19 epidemic management. In Artificial Intelligence for COVID-19
(pp. 11–29). Springer, Cham.
[77] Kent, J. (2020). Understanding the COVID-19 pandemic as a big data ana-
lytics issue. Erişim Tarihi, 11, 2020.
[78] Giordano, G., Blanchini, F., Bruno, R., et al. (2020). A SIDARTHE model of
COVID-19 epidemic in Italy. arXiv preprint arXiv:2003.09861.
[79] Chen, B., Shi, M., Ni, X., et al. (2020). Visual data analysis and simulation
prediction for COVID-19. arXiv preprint arXiv:2002.07096.
[80] Tátrai, D. and Várallyay, Z. (2020). COVID-19 epidemic outcome predic-
tions based on logistic fitting and estimation of its reliability. arXiv preprint
arXiv:2003.14160.
[81] Strzelecki, A. (2020). The second worldwide wave of interest in coronavirus
since the COVID-19 outbreaks in South Korea, Italy and Iran: a Google
Trends study. arXiv preprint arXiv:2003.10998.
[82] Long, Y. S., Zhai, Z. M., Han, L. L., et al. (2020). Quantitative assessment of
the role of undocumented infection in the 2019 novel coronavirus (COVID-
19) pandemic. arXiv preprint arXiv:2003.12028.
[83] Mahajan, A., Sivadas, N. A. and Solanki, R. (2020). An epidemic model
SIPHERD and its application for prediction of the spread of COVID-
19 infection in India. Chaos, Solitons & Fractals, 140, 110156.
[84] Heroy, S. (2020). Metropolitan-scale COVID-19 outbreaks: how similar are
they? arXiv preprint arXiv:2004.01248.
[85] Awotunde, J. B., Ogundokun, R. O., Adeniyi, A. E., Abiodun, K. M., and
Ajamu, G. J. (2022). Application of mathematical modelling approach in
COVID-19 transmission and interventions strategies. Studies in Systems,
Decision and Control, 366, 283–314.
[86] Zhao, X., Liu, X. and Li, X. (2020). Tracking the spread of novel cor-
onavirus (2019-nCoV) based on big data. medRxiv.
[87] Zhou, C., Su, F., Pei, T., et al. (2020). COVID-19: challenges to GIS with
big data. Geography and Sustainability, 1, 77–87.
[88] Jin, Y. H., Cai, L., Cheng, Z. S., et al. (2020). A rapid advice guideline for
the diagnosis and treatment of 2019 novel coronavirus (2019-nCoV) infected
pneumonia (standard version). Military Medical Research, 7(1), 4.
[89] Ahmed, S. F., Quadeer, A. A. and McKay, M. R. (2020). Preliminary iden-
tification of potential vaccine targets for the COVID-19 coronavirus (SARS-
CoV-2) based on SARS-CoV immunological studies. Viruses, 12(3), 254.
[90] Banerjee, A., Santra, D. and Maiti, S. (2020). Energetics based epitope
screening in SARS CoV-2 (COVID 19) spike glycoprotein by immuno-
informatic analysis aiming to a suitable vaccine development. bioRxiv.
[91] Ganasegeran, K. and Abdulrahman, S. A. (2020). Artificial intelligence
applications in tracking health behaviors during disease epidemics. In
Human Behaviour Analysis Using Intelligent Systems (pp. 141–155).
Springer, Cham.
[92] Ge, Y., Tian, T., Huang, S., et al. (2020). A data-driven drug repositioning fra-
mework discovered a potential therapeutic agent targeting COVID-19. bioRxiv.
[93] Enayatkhani, M., Hasaniazad, M., Faezi, S., et al. (2020). Reverse vacci-
nology approach to design a novel multi-epitope vaccine candidate against
COVID-19: an in silico study. Journal of Biomolecular Structure and

Dynamics, 39, 2857–2872.
[94] Koirala, A., Joo, Y. J., Khatami, A., Chiu, C. and Britton, P. N. (2020).
Vaccines for COVID-19: the current state of play. Paediatric Respiratory
Reviews, 35, 43–49.
[95] Petropoulos, G. (2020). Artificial intelligence in the fight against COVID-
19. Bruegel (23 March).
[96] Bullock, J., Pham, K. H., Lam, C. S. N. and Luengo-Oroz, M. (2020).
Mapping the landscape of artificial intelligence applications against
COVID-19. arXiv preprint arXiv:2003.11336.
[97] Harari, Y. N. (2020). The world after coronavirus. Financial Times, 20.
[98] Naudé, W. (2020). Artificial intelligence against COVID-19: an early
review.
[99] Maier, B. F. and Brockmann, D. (2020). Effective containment explains
subexponential growth in recent confirmed COVID-19 cases in China.
Science, 368(6492), 742–746.
[100] Huang, J. J. (2020). COVID-19 and applicable law to transnational personal
data: trends and dynamics. Sydney Law School Research Paper, (20/23).
[101] Chen, B. (2014). Historical foundations of choice of law in fiduciary obli-
gations. Journal of Private International Law, 10(2), 171–203.
[102] Douglas, M. (2018). Characterisation of breach of confidence as a privacy
tort in private international law. UNSWLJ, 41, 490.
[103] Alzahrani, N. M. (2020). COVID-19 pandemic: an overview of machine
and deep learning methods for analysis of digital media texts. Research in
Computing Science, 149(5), 179–189.
[104] “How DAMO academy’s AI system detects coronavirus cases,” (2020).
[Online]. Available: https://www.alizila.com/how-damo-academys-ai-sys-
tem-detects-coronavirus-cases/.
[105] Viboud, C., Sun, K., Gaffey, R., et al. (2018). The RAPIDD Ebola
forecasting challenge: synthesis and lessons learnt. Epidemics, 22,
13–21.
[106] Kahn, R., Mahmud, A. S., Schroeder, A., et al. (2019). Rapid forecasting of
cholera risk in Mozambique: translational challenges and opportunities.
Prehospital and Disaster Medicine, 34(5), 557–562.
[107] Armstrong, G. L., MacCannell, D. R., Taylor, J., et al. (2019). Pathogen
genomics in public health. New England Journal of Medicine, 381(26),
2569–2580.
[108] Schiavo, R., May Leung, M. and Brown, M. (2014). Communicating risk
and promoting disease mitigation measures in epidemics and emerging
disease settings. Pathogens and Global Health, 108(2), 76–94.
[109] De-Montjoye, Y. A., Gambs, S., Blondel, V., et al. (2018). On the privacy-
conscientious use of mobile phone data. Scientific Data, 5(1), 1–6.
[110] Ray, E. L. and Reich, N. G. (2018). Prediction of infectious disease
epidemics via weighted density ensembles. PLoS Computational
Biology, 14(2), e1005910.
[111] Cottle, M., Hoover, W., Kanwal, S., Kohn, M., Strome, T. and Treister, N.
(2013). Transforming Health Care Through Big Data Strategies for
Leveraging Big Data in the Health Care Industry. Institute for Health
Technology Transformation, http://ihealthtran. com/big-data-in-healthcare.
[112] Hotelling, H. (1933). Analysis of a complex of statistical variables into
principal components. Journal of Educational Psychology, 24(6), 417.
[113] Folorunso, S. O., Awotunde, J. B., Adeboye, N. O. and Matiluko, O. E.
(2022). Data classification model for COVID-19 pandemic. In Advances in
Data Science and Intelligent Data Communication Technologies for
COVID-19 (pp. 93–118). Cham: Springer.
[114] Dalal, N. and Triggs, B. (2005, June). Histograms of oriented gradients for
human detection. In 2005 IEEE Computer Society Conference on Computer
Vision and Pattern Recognition (CVPR’05) (Vol. 1, pp. 886–893). IEEE.
[115] Cohen, J. P., Morrison, P., Dao, L., Roth, K., Duong, T. Q. and Ghassemi,
M. (2020). Covid-19 image data collection: prospective predictions are the
future. arXiv preprint arXiv:2006.11988.
Chapter 3
Cloud-based IoMT for early COVID-19
diagnosis and monitoring
G. Boopathi Raja1, T. Sathya1, V. Gowrishankar1 and
M. Parimala Devi1
Abstract
COVID-19 has been posing a threat to survival from the second half of 2019. Each
country in the world must fight against the COVID-19 pandemic with caution. Many
researchers around the world have developed many vaccines, but some of them are
found to be effective. This chapter introduces an effective scheme for the diagnosis
and tracking of patients with COVID-19 based on symptoms. The concept behind the
proposed approach is to use an Internet of Things (IoT)-based system to handle the
real-time symptom data from patients in order to diagnose coronavirus cases early.
Additionally, the framework has the ability to track the medical records of those who
have healed from the COVID-19 disease. The proposed framework must learn
automatically about the origin of the virus by monitoring and analyzing necessary
data. Because of the significant and rapid rise in the number of patients after the
COVID-19 pandemic, it is crucial to focus an eye on patients’ health before any new
disease or infection occurs. IoT security has recently become a serious concern and a
hard problem. For researchers, transferring the large amount of collected healthcare
information data of patients who do not want their personal healthcare details shared
has remained a difficult task. The health status of patients is determined in this model
by predicting critical situations and examining physiological data received from
smart medical IoT devices, ensuring that patients’ personal information is kept pri-
vate. Based on current advancements, the suggested model is thought to be
suitable for delivering an appropriate remote patient monitoring model with accurate
data in cloud-based IoT systems.
Keywords: COVID-19; Early diagnosis; Internet of Medical Things (IoMT);

Disease prediction; Real-time health tracking
1
Department of Electronics and Communication Engineering, Velalar College of Engineering and
Technology, Erode, India
3.1 Introduction
As of May 27, 2021, more than 168 million COVID-19 confirmed cases had been
recorded worldwide since its detection in late December 2019, with a daily rise of
about 5%. More than 95,000 people have died as a result of these incidents,
representing a 4.2% mortality rate. The World Health Organisation classified this
unique coronavirus as a pandemic on March 11, 2020. Unfortunately, there is no
effective therapy or vaccination available at this time. The production of an
effective vaccine was done, and it had taken over a year, especially because the
identification of the existence of the virus is a challenging task. However, the
production of the number of vaccines to the entire population worldwide is not
sufficient due to a lack of resources [1].
Coronaviruses are a broad category of viruses with a wide range of char-
acteristics. Some of these make sick people with colds. Others infect animals such
as cattle, camels, and bats. The first human coronavirus was found in the year
1965. It was the cause of a common cold. During the decade, researchers
uncovered a set of identical human and animal viruses with crown-like appear-
ances [2].
Seven coronaviruses are capable of infecting humans. In 2002, the deadly
SARS virus was identified in southern China and rapidly spread to 28 other
countries. More than 8,000 persons had been affected by July 2003, with 774 of
them dying. Only four additional cases were reported in a minor epidemic in 2004.
Fever, headaches, and respiratory problems such as coughing and breathing pro-
blem are all symptoms of the coronavirus [3].
In 2012, MERS was discovered for the first time in Saudi Arabia. Nearly 2,500
cases are found among the people who stay in or visited the Middle East recently.
The coronavirus, though less contagious than SARS, is more deadly, having killed
858 people. It involves respiratory problems similar to a regular cold, but it can also
lead to renal failure.
Currently, the only method in the world for dealing with this coronavirus is to
use social isolation, hand washing, and face masks to limit its spread. By allowing
for early identification and tracking of new outbreaks, technology, on the other
hand, could be able to help limit the disease’s spread. These approaches require
cloud, big data, and fog capabilities, as well as the handling of information col-
lected via remote monitoring to track the real-time health status of the patients
located at long distances, teleHealth, m-Health [4]. This research presents an
effective approach for the diagnosis and tracking of COVID-19 patients that uses
wearable sensor technology to capture real-time symptom data.
The summary of this chapter is discussed as follows. Section 3.2 elaborates the
symptoms of COVID-19, possible diagnostic methodologies, therapeutic techniques,
suitable vaccination along the important happenings worldwide as a COVID-19
timeline. The existing works along with the requirements are described under
Section 3.3. Section 3.4 introduces the methodology and architecture of the proposed
framework along with a flow diagram. Section 3.5 describes the implementation of
the proposed framework cloud-based IoMT system for early COVID-19 diagnosis
Cloud-based IoMT for early COVID-19 diagnosis and monitoring 63
and monitoring. Section 3.6 demonstrates the results and discussion with
suitable illustrations.
3.2 Overview about COVID-19 treatments
3.2.1 Symptoms
COVID-19 has a wide range of impacts on different people. The majority of sick
people will have minor to severe symptoms and will heal without needing to go to
the hospital. Patients having COVID-19 have a different range of problems ranging
from mild to moderate pain and sometimes up to life-threatening sickness. COVID-
19 has a wide range of impacts on different people. The majority of sick people will
have minor to severe symptoms and will survive without requiring even going to
the hospital. Symptoms include nausea or vomiting, severe diarrhea, and a new lack
of taste or odor [5].
Fever, dry cough, and fatigue are the most common symptoms in COVID
patients. Any of the less common symptoms are aches and pains, indigestion, fever,
loss of color in fingers, lack of taste or odor, a rash on the scalp, conjunctivitis, and
sore throat. COVID-19 symptoms include tiredness, difficulties in movement, and
shortness of breath or difficulty in breathing, chest pain, and loss of speech.
Patients who are experiencing severe symptoms require prompt medical
intervention. Mild symptoms should be managed at home by otherwise healthy
people. When a patient is infected with the virus, signs usually start after 5–6 days,
although it can take up to 14 days.
COVID-19 illness is more likely to have major symptoms in the elderly and
those with severe chronic medical problems such as heart disease or diabetes.
Patients experiencing problems in breathing, persistent discomfort or pain in
the chest, gray, pale or blue-colored skin, lips, or nail beds, depending on skin tone,
and an inability to wake or stay awake should seek immediate medical assistance.
3.2.2 Methodologies in COVID-19 diagnosis

SARS-CoV-2 can currently be diagnosed using one of two methods: detection of
the antibody produced after infection or detection of viral RNA. The most prevalent
methods for detecting SARS-CoV-2 viral RNA are polymerase chain reaction
(PCR) and nucleic acid hybridization. Viral antibodies or antigens can be detected
using ELISA and other immunological and serological methods [6]. The identifi-
cation of the virus in its active state is accomplished by determining the virus’s
RNA, while serological assays aid in identifying people whose immune systems
have already formed antibodies to combat the infection.
● Amplification techniques
– Isothermal nucleic acid amplification
– RT-PCR (reverse transcription-PCR)
● Immunological assays
● SARS-CoV-2 diagnosis using newly discovered technology
For the time being, the most widely used procedures for diagnosing COVID-19
are RT-PCR and immunological testing. These procedures, however, necessitate
the use of qualified staff. Furthermore, the findings of PCR might take up to a few
days to get. Antibodies and recombinant proteins are required for immunological
tests. As a result, there is a tendency toward developing new SARS-CoV-2 diag-
nostic procedures that are quicker, less expensive, and more reliable.
The findings of RT-PCR can be received in a few hours and have good sen-
sitivity and specificity. In addition to lung samples, saliva, blood, urine, and stools,
it may identify viral DNA. However, there are significant disadvantages to RT-
PCR, including the requirement for a costly thermocycler and experienced
employees to execute the experiment and interpret the data. Furthermore, bench-
mark validation is critical for the validity of the data, as false-negative findings
might occur as a result of sample deterioration, sample collection quality and time,
and inefficiency of some test kits. The sensitivity and specificity of LAMP tech-
niques are equivalent to those of RT-PCR. Some kits, on the other hand, have
reduced sensitivity. It can be done for about half an hour with a crude sample,
allowing them to be used in POC testing. SARS CoV-2 detection has also been
created using the CRISPR approach, which has great sensitivity and specificity. It
takes 1 h to complete and can be used with the lateral flow assay [7]. For LAMP
and CRISPR, a costly thermocycler is not required. A lateral flow assay is a simple
approach that may be used by nonprofessionals to produce findings in 15 min in
blood or serum samples. Storage, shipping, and sample collection do not affect
antibodies. Its disadvantage is that it takes a long time to manufacture antibodies.
Although ELISA is simple to use, it cannot be utilized for early detection like the
lateral flow test. It can, however, be used to test the immunity of healthcare workers
and to investigate herd immunity.
3.2.3 Treatment approaches

The number of medicinal options available to treat COVID-19 is continually
increasing and developing. Clinicians, patients and their representatives, and health
system management are highly encouraged to study the COVID-19 Treatment
Guidelines issued by the National Institutes of Health frequently, according to the
CDC [8].
Infection prevention and control methods, as well as supportive care, such as
supplementary oxygen and mechanical ventilator support, are being used in the
treatment of COVID-19. In some cases during the treatment of COVID-19, the
FDA has been permitted to use authorized remdesivir (Veklury) based on the health
status of the patient [7,9].
Early and effective treatment of any disease can assist forestall the advancement
of the disease to a more serious condition, especially in individuals who are at high
risk of disease progression and severe illness, while also minimizing the load on
healthcare systems. Under the EUA, a variety of new treatments (such as monoclonal
antibodies) are accessible for early outpatient therapy. Outpatients with a high risk of
illness progression are undergoing trials to see if these treatments are successful.
3.2.4 Available vaccine

As of December 2020, there are over 200 COVID-19 vaccine candidates under
development. At this time, at least 52 candidate vaccinations are being tested in
humans. Several more are in phase I/II right now and will go to phase III in coming
months [9].
A significant number of people are typically tested before a vaccination can-
didate is shown to be both safe and efficacious. Around 7 out of every 100 vaccines
tested in the lab and on laboratory animals will be approved for human clinical
trials. Only one out of every five vaccines tested in clinical trials is successful.
When a large number of vaccinations are being developed and tested, the chances
of one or more successful vaccines proving to be safe and effective for specific
population rises.
Vaccination can be designed in one of three ways. They differ in whether they
use the entire bacterium or virus, just the bits that activate the immune system, or
just the genetic data that direct the production of specific proteins rather than the
entire virus.
1. The whole-microbe approach
2. Viral vector vaccine
3. The subunit approach
3.2.4.1 The whole-microbe approach

Inactivated vaccine
The initial stage in creating a vaccine is to use chemicals, heat, or radiation to neu-
tralize or eliminate the infectious virus, or a virus that is genetically identical to it.
This method employs technology that has been shown to function in humans, such as
flu and polio vaccinations, and pharmaceuticals may be mass produced at a low cost.
However, expensive equipment is necessary to adequately develop the virus or
bacteria, and the procedure might take a long time. Two or three treatments are
nearly always necessary.
Live-attenuated vaccine
A virus that seems to be alive and has been diluted, or one that is highly close, is
used in a live-attenuated vaccine. Rubella, mumps, and measles vaccinations, as
well as chickenpox and shingles vaccinations, fall under this category. This process
is equivalent to the production of neutralizing antibodies and may be done in huge
quantities. People with weakened immune systems, on the other hand, may not be
able to benefit from such vaccinations.
3.2.4.2 Viral vector vaccine

This type of immunization uses a harmless virus to transfer specific proteins from
the germ of interest to trigger an immune response without causing sickness. The
protein is subsequently transported into the body by the virus, which functions as a
carrier for it. The protein is the catalyst for the immune response. The ebola vac-
cines are being manufactured quickly since it is a viral genome vaccine.
3.2.4.3 The subunit approach

Only the fragments of a virus or bacterium that the immune system needs to detect
are used in a subunit vaccination. It does not have a harmless virus as a vector, and
it does not contain the full microbe. Proteins or carbohydrates might be used as
subunits. The majority of vaccinations on the pediatric immunization schedule are
subunit vaccinations that protect against illnesses, including diphtheria, tetanus,
whooping cough, and meningococcal meningitis.
The genetic approach (nucleic acid vaccine)
A nucleic acid vaccine, unlike all other vaccinations that employ the complete
organism or sections of it, employs simply a piece of genetic material that offers
instructions for certain proteins, instead of the complete bacteria. In the form of
mRNA or DNA, a nucleic acid vaccine gives the cells particular instructions to
build the protein that the immune system needs to detect and respond to.
The nucleic acid method is a novel technique for vaccine development. None
had gone through the whole licensing procedure for use in people before the
COVID-19 pandemic; however, certain DNA vaccines, particularly those for spe-
cific tumors, were in human testing. As a result of the pandemic, investigation in
this area has increased, and many COVID-19 mRNA vaccines have been granted
emergency use authorization, allowing them to be delivered to people who are not
in clinical trials.
3.2.5 COVID-19 timeline

Table 3.1 describes the important highlights that have been happened from
December 2019 in the Wuhan city of China to date around the world [10,11].
3.3 Related work

3.3.1 Lightweight block encryption–based secure health
monitoring system for data management
Akhbarifar et al. have suggested the encrypted secure customized healthcare
approach in a cloud-based Internet of Things (IoT) context. As a result, the major
purpose of this study is to develop a credible health monitoring model for an early
detection of HD (heart disorders), HTN (hypertension), and HCLS (hypercholes-
terolemia) by forecasting the critical patient’s state utilizing the procedures as
follows [12]:
1. Medical IoT devices collect biological data from patients for remote medical
monitoring.
2. Using a suggested lightweight block encryption approach to safeguard medical
IoT data and ensure the security and confidentiality of patient medical data.
3. Using the cloud to transfer encrypted data for disease prediction.
4. Detecting unexpected variations in blood cholesterol levels in patients and
predicting the HCLS.
Table 3.1 COVID-19 timeline
December 31, 2019 The Wuhan Municipal Health Commission has confirmed a cluster of
pneumonia cases in Hubei Province, China. A new coronavirus was
later discovered
January 4, 2020 A group of pneumonia cases was reported by the WHO on social
networking sites in Wuhan, Hubei province, although no deaths
were reported
January 12, 2020 COVID-19’s genetic sequence was published by China
January 13, 2020 The first occurrence of COVID-19 outside of China was confirmed in
Thailand
January 20 and 21, A short field visit to Wuhan was performed by WHO specialists from
2020 the China and Western Pacific regional offices
February 3, 2020 The International Community’s Strategic Preparedness and Response
Plan, published by the WHO, would help in the security of countries
with poor health systems
February 11 and 12, Over 400 experts and key contributors from across the world attended
2020 a COVID-19 Research and Innovation Forum, which featured
lectures by experts from CDC
March 19, 2020 ● All incoming international flights banned in India
● Export of masks, ventilators and certain medications are banned by
Indian government
March 20, 2020 ● WHO delivered 1.5 Million Lab test kits to many countries
● The largest single increase of the corona virus outbreak recorded
627 deaths in Italy

March 22–25, 2020 ● Global corona virus cases doubled and recorded 330,000 cases
● WHO declared the “pandemic is accelerating”
● Almost one-third of world population is affected by corona virus

lockdown
● “There will be a shortage of medical supplies” announced by WHO
April 2, 2020 Global corona virus cases recorded the one million; mortality exceeds
50,000
April 8, 2020 Global corona virus cases surpass 1.5 million
April 22, 2020 First trials for a corona virus vaccine are approved by Germany
May 3, 2020 The total number of fatalities in the United States was 65,464 people
May 7, 2020 The UN reports that the corona virus has affected 14 million people.
The mortality rate in the United States has reached 1 lakh
May 28, 2020 Latin America accounted for about 40% of COVID-19 mortality
worldwide
June 11, 2020 In July, the Biotech company Moderna began the last phase of a
testing for a corona virus vaccine on 30,000 people
June 23, 2020 The region’s first vaccination trial has begun in South Africa. At the
same time, the number of people who have died as a result of the
corona virus in Latin America has reached 1 lakh
June 24, 2020 India reports over 15,000 and it was the highest single-day rise new
corona virus infections
July 2, 2020 In spite of corona virus cases surpass 600,000 and 17,834 deaths, India
announces reopening measures
July 16, 2020 India records 1 million corona virus cases. The corona virus outbreak
rises in rural areas of the country
(Continues)
Table 3.1 (Continued)
August 1, 2020 More than 57,000 new victims were reported in India
August 13, 2020 COVID-19 cases have increased by over 67,000 in India. It appears
that this is the largest daily rise since the epidemic began
August 15, 2020 Russia starts production of vaccine: Sputnik V COVID-19
September 4, 2020 India has reached 83,000 new positive victims
September 5, 2020 India currently has 90,000 daily cases, overtaking Brazil as the country
with the second-highest number of cases
October 12, 2020 After a volunteer experience an unexpected illness, Johnson &
Johnson suspends its COVID-19 vaccination trials
October 29, 2020 COVID-19 cases have reached 8 million in India
November 20, 2020 ● The world passes 50 million corona virus cases
● Preliminary data suggest that Pfizer and BioNTech’s vaccine is
more than 94% effective

December 21, 2020 According to the WHO, the UK variety of SARS-CoV-2 looks to be
more contagious, although vaccinations are unlikely to be affected
December 23, 2020 The first 1 million COVID-19 vaccinations are given out in the United
States
February 4, 2021 ● COVID-19-related fatalities in the United States have exceeded
450,000, while COVID-19-related fatalities in Spain have
crossed 60,000
● Johnson & Johnson has requested emergency use permission for its
COVID-19 vaccine from the FDA
● The vaccine developed by Oxford-AstraZeneca is effective against
the United Kingdom corona virus strain
March 4, 2021 ● India’s Covaxin COVID-19 vaccine was found to be around 80%
effective in a new study
● Germany approves AstraZeneca’s COVID-19 vaccine
March 12, 2021 The vaccine developed by Johnson & Johnson has been approved by
the WHO for emergency use
April 20, 2021 According to the WHO, worldwide COVID-19 cases have risen for the
seventh week in a row. The number of people who have died as a
result of the coronavirus has exceeded 3 million worldwide
May 1, 2021 In a single day, India becomes the first country to report over 4 lakh
cases
May 28, 2021 ● Globally, the total confirmed cases of COVID-19 have reached over
169,021,000, with more than 3,511,000 deaths and over
118,593,000 recoveries
● The total number of COVID-19 cases has crossed 30 million in
South Asia
● India is the most heavily affected country in Asia, but neighboring
countries such as Nepal, Bangladesh, and Pakistan are also suffering
from the disease
5. Detecting HD in the event of an HTN diagnostic test and predicting HTN risk
and severities.
6. Providing the medical teams with the disease prediction process’s analytical
results.
This secure health-monitoring approach is separated into four elements, as

explained next.
3.3.1.1 IoT network and data collection

This section contains medical IoT sensors, network devices, and tools for detecting
and collecting physiological data from patients. The patient’s vital signs, such as
heart rate, blood pressure, blood cholesterol, and other essential biological data, are
acquired via the body area network by sensors implanted on the patient’s clothing
or body. A security requirement for secure IoT data has been included because
medical IoT sensor network devices are more vulnerable to security attacks than
conventional network devices [13,14]. The IoT data are secured using a lightweight
block encryption approach before being sent to the clouds with the obtained
medical data.
3.3.1.2 Communication service provider

This section is in charge of sending the medical data collected from the patients to
the cloud storage [15]. This component must offer confidential sharing for trans-
mission to cloud servers as part of a distributed data storage framework.
3.3.1.3 Distributed data storage

This section stores the medical data that have been transmitted by medical IoT
sensors. Also included in the distributed data storage component is the provisioning
and delivery of services to the interested users, which include doctors and health-
care professionals. These capabilities might be incorporated into a facility that uses
data-mining techniques to estimate the probability of illness. The suggested safe
health-monitoring model takes into account a combination of three categories of
associated illnesses, including hypertension, hypercholesterolemia, and a heart
condition.
3.3.1.4 Healthcare provider

Doctors, hospitals, and emergency responders make up this section. Doctors can
use the forwarded diagnosis results to double-check and confirm them before
making necessary medical recommendations to patients.
3.3.2 Smart diagnostic/therapeutic framework for COVID-

19 patients
The COVID-19 monitoring unit is used to identify suspected people who have
primary symptoms. The clinical data of individuals with mild symptoms who might
be in a self-quarantine/home environment with few difficulties are tracked utilizing
cloud technology. While obtaining those data, the government’s safety rules are
followed. The adoption of this supporting and tracking system creates a peaceful
situation for COVID-19 patients’ families and friends [16,17].
These systems also assist the health workers in providing medications when-
ever required. It aids healthcare providers in not just monitoring patient information
but also storing it for future investigation. Patients’ essential information is
continually recorded without being contacted. This system has a minimum of four
primary sensors to update patient’s data regularly. This information is updated on
mobile phones for monitoring purposes.
Monitor respiration rate, temperature, and pulse of a patient are a major con-
cern and hence the data acquired by the signal conditioning unit could be connected
to mobile phones of the patients. The system monitors COVID-19 patients in real
time rapidly. The efficiency of the monitoring system has improved by the data
collected from the sensors, which are saved in the cloud also in a regular period.
This system not only helps the suspected COVID-19 patients but also helps
healthcare providers to monitor many patients in remote places.
3.3.3 IoT-based framework for collecting real-time symptom

data using machine learning algorithms
The research was carried to explore if machine learning techniques might be used
to swiftly detect (or predict) likely COVID-19 diseases [18–21]. The remaining
sections of this section go through the experimental setup as well as the results that
are presented and discussed.
3.3.3.1 Dataset
The COVID-19 Open Research Collection (CORD-19) repository has made a
collection of 14,251 validated COVID-19 cases available. The data in each case
comprises a variety of data types. Travel history to suspect regions, symptoms, and
contacts with potentially infected people were all investigated. However, for many
of the occurrences listed in the database, part of this information was incomplete.
Furthermore, the data was not adequately formatted for machine learning algo-
rithms to work with.
3.3.3.2 Data preprocessing

To make it easier for machine learning, the data was preprocessed and formatted.
We gathered all of the instances that had recorded symptoms. A total of 80 symp-
toms emerged as a result of this. Many of these symptoms, however, were deter-
mined to be interchangeable. As a result, there were just 20 symptoms left. This ad
hoc merging of identical symptoms was done by two medical practitioners. For
example, the terms “anorexia” and “lack of appetite” were combined.
The relative relevance of these 20 symptoms was also assessed in prior
research. The important key parameters for feature selections have been listed as
follows: correlation analysis, interquartile range, intraclass distance, spectral score,
information score, and variance-based feature weighting. The variance-based fea-
ture weighting method is one, in which, it not only rates the symptoms but also
gives each one a weighted significance rating. Cough, sore throat, fever, fatigue,
and shortness of breath were discovered to be the most critical five symptoms
(ranked from most essential to least significant).
This study employs the five most essential symptoms based on the findings of
the previous study. There were also two new features added: contact and live. The
first attribute (touch) specifies if the person has previously been known to have
come into contact with a potentially confirmed person. The second characteristic
(live) indicates whether the person resided, traveled to, or passed through a con-
taminated location. As a result, a 14,767-data-record preprocessed dataset was
produced. There were 854 COVID-19 cases whose records had been confirmed and
622 COVID-19 cases whose records had not been confirmed.
3.3.3.3 Predictive model

The preprocessed data was utilized to develop a prediction model for our classifi-
cation method in this study. The goal of this model is to forecast how likely a
person is to become infected with COVID-19.
This might have been accomplished using a variety of learning techniques (i.e.
classifiers). These classifiers can be divided into several groups. The classifiers are
divided into six groups by WEKA Software, which we employed in this study:
1. Lazy classifiers, such K-nearest neighbors (K-NN).
2. Meta classifiers, such neural networks.
3. Function-based classifiers, such as support vector machines (SVMs).
4. Tree-based classifiers, such as decision stump.
5. Bayes-based classifiers, such as naı̈ve Bayes.
6. Rule-based classifiers, such as Decision Tables, ZeroR, and OneR.
All these algorithms were conducted on the dataset using WEKA Software. For
each of the eight algorithms, the default parameter values were utilized. The eight
algorithms are described briefly in the following:
Support vector machine (SVM). SVM trains the hyperplane that divides
the instances from each class while also increasing the distance between the data
instances and the hyperplane, given a collection of labeled training examples.
The learned hyperplane is then applied to every new test object to assign a
class label.
Artificial neural network (ANN). The learning process tries to recreate what
happens in the human brain when it comes to learning. To do this, edges connect
many layers of nodes. Numerical weights are used to represent the edges linking
the nodes. Each node’s output is calculated as the weighted sum of its inputs.
The ANN learns the statistical weights that identify the instances for every
class consist of a set of labeled training images (each case corresponds to either the
positive or negative class). Each test case is then assigned a class label using the
trained model. The outputs of the last layer are presented to a threshold to deter-
mine the label with this test instance.
Naı̈ve Bayes. Naı̈ve Bayes is one of the supervised learning approaches. A
probabilistic technique is used in the learning process. The model parameters are
computed using the Bayes theorem.
Naı̈ve Bayes creates several parameter values, such as the likelihood of every
class label occurring, given a collection of labeled training data (each case corre-
sponds to either the positive or negative class). Following that, each test case is
assigned a class label based on these variables. This is done by estimating the
likelihood of the test instance being matched to each of the possible attribute
values. The label of that test case is determined by the largest value among these
probabilities.
K-Nearest neighbors (K-NN). It belongs to supervised learning. The process
of learning moves at a slow pace. There is no model creation.
K-NN calculates the similarities among a given test instance along with all the
training instances for a specific set of labeled training examples (each case corre-
sponds to either the positive or negative class). The test instance’s class label is
then assigned (or predicted) using these distances. The class names of the K
training samples closest to the test sample are combined to achieve this.
Decision table. Another example of a supervised learning approach is the
decision table. In this technique, it generates a decision table from a set of labeled
training cases to calculate a classifier (each case corresponds to either the positive
or negative class). The table is made up of a series of criteria and actions. It is said
to be complete when it assesses all combination cases for the criteria and provides
appropriate steps with each of them.
Decision stump. This technique belongs to supervised learning. In this tech-
nique, it creates a decision tree for only one input feature to find a model from a
group of labeled training instances. Alternatively, it makes a forecast for each test
instance based just on one sample attribute. Each component is selected by calcu-
lating the entropy of all features across all training samples and choosing the one
with the highest accuracy.
Zero rule (ZeroR). These types of algorithms fall under the category of
supervised learning. Based on the group of labeled training cases, this algorithm
defines a classifier using just the target feature (i.e. class) and neglecting most of
the other features. It belongs to the basic classification approach. The majority class
is assigned to any new test instance. It is usually utilized as a starting point for
determining baseline performance.
One rule (OneR). Another technique that falls under supervised learning is
OneR. This technique generates one rule for every feature in the dataset from a set
of labeled training instances (each case corresponds to either the positive or
negative class). After that, the one with the smallest overall error is selected.
3.4 Proposed methodology

The concept behind the suggested technique is to gather real-time symptom data
from patients using an artificial intelligence (AI)-based framework to identify cor-
onavirus infections in the early stages. Furthermore, the AI-based framework can
track the medical records of people who have recovered from the COVID-19 dis-
ease. By obtaining and evaluating relevant information, the proposed approach must
automatically learn about the COVID-19 spreading mechanism. The system’s five
major components are diagnostic data collection and sharing, quarantine/isolation
center, predictive analytics center, health specialists, and cloud infrastructure. Both
AI-based smart medical devices and IoT have altered health-monitoring services by
allowing patients’ health conditions to be tracked and operated remotely at any time
and from any location. Because of the significant and rapid rise in the number of
patients after the COVID-19 pandemic, it is crucial to focus on patients’ health
before any serious symptoms occur.
3.4.1 Architecture of proposed IoT framework

This chapter explains the IoT-based proposed architecture for real-time monitoring
and identification (or prediction) of COVID-19-infected persons. This framework
might aid in improving the treatment process for reported cases and gaining a better
knowledge of the nature of the COVID-19 illness.
The architecture of the suggested IoT-based framework is shown in Figure 3.1.
It is consists of five major elements as follows:
● Collection and upload of symptom data using smart medical IoT device along
with Google assistant.
● A center for quarantine and isolation for COVID-19 patients.
● A data analysis center—a place where people may go to get their data
analyzed.
● A way to connect with doctors using IoT health management.
● The cloud infrastructure connects them all.
Cloud
Physiological Uploading
parameters patients data
Consultation and Retrieving

suggestions required data
Collection and uploading Health experts team Data

of symptom data analysis center
e
antin
Quar
Centre
for quarantine and isolation
Figure 3.1 Architecture of proposed AI-based framework

3.4.1.1 Collection and upload of symptom data

The objective of this component is to use Smart IoT wearable sensor devices along
with Google assistant to record real-time symptom reports from the patient’s body.
Based on an actual COVID-19 patient dataset, we determined the most significant
COVID-19 symptoms in a previous study. Sore throat, fatigue, cough, fever, and
shortness of breath were among the symptoms noted.
These major symptoms can be detected using biosensors IC modules in the IoT
device. Shortness of breath, for example, may be detected using oxygen-based
sensors. Fever may be detected using temperature-based sensors. Fatigue may be
detected using both heart-rate and motion-based sensors. The image-based cate-
gorization can be used to detect a sore throat. Audio-based sensors combined with
acoustic and aerodynamic models could detect and classify cough.
Other pertinent information, such as contact and travel history from the pre-
vious few weeks, may be obtained on the move using the Aarogya Setu app.
3.4.1.2 Center for quarantine and isolation

People who have been isolated or confined in a medical facility provide informa-
tion to this component. These documents include both medical and nonmedical
data. Each record contains data information with the previously mentioned symp-
toms for health data, and each record also includes nontechnical data such as travel
and contact history over the previous 3–4 weeks, gender, age, chronic diseases,
genetic problems, and any other medical history. Each patient’s treatment reaction
would eventually be recorded in their records. The patients update their health
status using Google assistant themselves.
3.4.1.3 Data analysis center—a place where people may go to

get their data analyzed
COVID patient health records are analyzed at the data center. AI-based algorithms
are used to generate a COVID-19 model and a real-time interface for the data that
has been analyzed. Based on the information obtained and submitted by patients,
the algorithm may subsequently have been used to detect or anticipate probable
COVID-19 incidents. The patient’s treatment response may also be predicted using
the model. Disease models built from this evidence will eventually disclose
important knowledge regarding the nature of the sickness.
3.4.1.4 Way to connect with doctors

Physicians will focus on any potential infections that our suggested AI-based identi-
fication/prediction model advises inspired by real symptom data. Physicians should be
able to act quickly to these verified instances by doing any necessary medical tests.
This makes it possible to isolate confirmed cases and treat them properly.
3.4.1.5 The cloud infrastructure connects them all

The IoT cloud platform permits users to upload real-time symptom data, keep
medical records, share prediction findings, engage with specialist recommenda-
tions, and save data.
Collecting Centre
and uploading of symptom data for quarantine and isolation
Fever, cough, fatigue, sore throat and
shortness of breath are examples of Maintaining digital records of those who
relevant symptoms that may be collected have been placed in quarantine or
using wearable devices. In addition to isolation, including confirmed and
user data such as travel and contact suspected cases
information.
Data analysis center—

a place where people may go to get their data
analyzed Continue to learn new approaches
utilizing digital data from health-care
providers and users.
Suspected cases are identified and
predicted using appropriate algorithms or
techniques.
Analyzing and monitoring the collected

reports
Health experts
Suspected cases should be monitored and Contact suspected cases and medical
reviewed facilities.
Figure 3.2 Stages of proposed framework
The scenario of the framework (or process) is depicted in Figure 3.2, which
may be summarized as follows.
The system uses smart biowearable sensors to collect noninvasive real-time
user symptom data. Cough, sore throat, fatigue, fever, and shortness of breath are
the symptoms again. Furthermore, the user enters information regarding living in
(or traveling to) contaminated locations, as well as probable interaction with
COVID-19-affected people, using a mobile application. Regularly, the quarantine/
isolation center delivers statistics on isolated and confined patients who are held
there. That information is analogous to the real-time information obtained from
patients.
Using Google assistant, the collected symptom data is sent to the cloud data
analysis center. Regularly, the smart biowearable sensor sends data to the health-
care center for data analysis. The data analysis center is home to AI algorithms that
update their models daily using data from the research center. Based on each user’s
real-time symptom data, the AI classifiers are then used to identify possible
instances. The data center’s data is evaluated in real time, and the results are dis-
played on a dashboard. The dashboard may be used by doctors to understand more
about the virus’s characteristics.
If a possible case is detected, the relevant physician will be alerted and the
patient will be immediately followed. After that, the patient will be called and
invited to come in for clinical testing, such as a PCR test to detect positive
instances. If the infection is confirmed, the victim will be quarantined and sepa-
rated from his or her contacts.
The use of the identical mobile app to communicate with consumers, deliver-
ing critical information about how to avoid illness and then being treated with
antibiotics, is an additional and critical component of this strategy.
3.4.2 Data acquisition using wearables devices

Figure 3.3 describes the workflow of the proposed framework. The procedure
starts from symptom data collecting from the patients through the sensors
attached with each patient. Then, the collected data is protected from intruders
by applying a suitable data encryption algorithm. The secured data of each
patient is allowed to store in the cloud and also transferred to the health experts.
The secured data can be analyzed by the experts after the appropriate data
decryption method. Based on command by experts on the data provided, the
further procedure to be followed effectively.
3.5 Implementation of proposed framework

Google Assistant is an AI power voice command service. We can engage with the
Google Assistant by speaking to it, and it can search the Internet, plan events,
set alarms, and operate appliances, among other things. This service can be
accessed by smart IoT biosensor devices. The Google Assistant conveys the patient
illness status to Health management systems. For this research Adafruit, IoT cloud
server is used as data analysis [22].
Figure 3.4 describes the implementation of the proposed framework to
obtain early diagnosis and provide suitable treatment for COVID-19 patients
using Google Assistant and Adafruit IoT cloud server. In this, framework IFTTT
(If This Then That)-based web service is used to interface Google Assistant with
the cloud platform. Sensor-based setup is attached with each patient who is
enabled by a smartphone. The health status of each patient is continuously
tracked through Google Assistant by experts/doctors who are placed at the
remote location. The Google Assistant may be handled by patients or family
members. IFTTT is a free web-based tool that allows users to construct applets,
which are chains of basic conditional statements. Changes in other online ser-
vices, such as Gmail, Facebook, Telegram, Instagram, or Pinterest, might trigger
an applet service [23].
Figure 3.5 shows the initial setup in the IFTTT web service. All the patients
are categorized into three groups. They are patients with critical conditions,
Data collecting from IoT device and

wearable sensors attached with
COVID-19 patients
Perform data encryption
Transferring and storing secured medical

data to the cloud
Distributing the data on request by

authorized person
Perform data decryption on secured data
Data preprocessing
Analysis using machine learning

algorithms
Normal Abnormal
condition condition
Confirmation of negative cases Confirmation of positive cases by

Diagnosis results
by consultation with doctors consultation with doctors
Yes No
End of treatment procedure Is the patient
critical?
Emergency Non-emergency
case case
Forwarding the diagnostic Forwarding the
results to the patients, family diagnostic results to
members and informing to the patients
emergency service providers
Provide immediate
Allow them to home
hospitalization to the patient
for continuous monitoring and quarantine
treatment
Figure 3.3 Flowchart of proposed framework
patients with normal symptoms (moderate), and patients with light symptoms or
no symptoms. Based on suggestions provided by experts, the keywords asso-
ciated with each symptom are grouped in each category.
Based on the command provided by the patient through the Google
Assistant, this setup automatically maps that patient to the corresponding
patient group. Through this, doctors can be able to give treatment on a
priority basis.
Cloud storage/
Connectivity via analytics
Patient with high fever IFTTT
Patient 1
Connectivity via
IFTTT
Google assistant
Connectivity via
Patient having mild
IFTTT
symptoms
Patient having no
Patient 2
symptoms Data analytics center
and hospital
Patient 3
Google assistant
Google assistant
Status of each patient can be tracked
continuously from cloud using
adafruit application
Figure 3.4 Implementation of proposed methodology using Google Assistant and

Adafruit API
Figure 3.5 Initial setup with IFTTT
3.6 Results and discussion
COVID-19 has a wide range of impacts on different people. The majority of sick
people will have minor to severe symptoms and will survive without requiring even
going to the hospital. Symptoms include nausea or vomiting, severe diarrhea, and a
new lack of taste, tiredness, difficulties in movement, and shortness of breath or
difficulty in breathing, chest pain, and loss of speech.
Figure 3.6 shows Google Assistant response for a person having cough
symptoms. Based on symptoms (cough) provided, the immediate response will be
provided to the patient. Figure 3.7 shows a Google Assistant response for a person
having no symptoms.
Patients who are experiencing severe symptoms require prompt medical
intervention. Mild symptoms should be managed at home by otherwise healthy
people. When a patient is infected with the virus, signs usually start after 5–6 days,
although it can take up to 14 days. COVID-19 illness is more likely to have major
symptoms in the elderly and those with severe chronic medical problems such as
heart disease or diabetes.
Figure 3.8 shows Google Assistant response for a person having fever and
Figure 3.9 shows Google Assistant response for a person having severe fever.
Based on the status provided, immediate actions will be taken.
Figure 3.6 Google Assistant response for person having cough symptoms
Figure 3.7 Google Assistant response for person having no symptoms
Figure 3.8 Google Assistant response for patient having fever

Figure 3.9 Google Assistant response for patient having high fever symptoms
Figure 3.10 illustrates that the cumulative reports of all the patients under
treatment are listed along with time. This report is continuously tracked and ana-
lyzed by health experts for further decisions.
Due to the increasing elderly population and persons suffering from life-threatening
chronic conditions, as well as the high expenses of care for all of these patients, the
rising demand for the health-monitoring framework required for the patients at
remote places has become a prime requirement in the present scenario in terms of
the coronavirus (novel COVID-19) pandemic. Real-time monitoring and analysis
of a patient’s health status can show crucial and abnormal circumstances, which
can aid in the early detection of any potentially fatal ailment. Advanced technol-
ogies along with cloud resources and medical IoT devices are critical in the
development of digital remote health-monitoring systems.
Figure 3.10 Cumulative reports of all the patients through Adafruit API
The healthcare services provided by the proposed framework can be extended

for providing treatment to various diseases to reduce the transportation cost, rental
expenses, and waiting period. Based on the need, the consultation will be provided to
the patients at right time with reasonable expenses. The various experts may interact
for each case and provide valuable suggestions as well as recommendations for each
healthcare service provider in an effective manner.
References
[1] Parimaladevi M., Sathya T., Gowrishankar V., Boopathi Raja G. and Nithya
S. An Efficient Control Strategy for Prevention and Identification of
COVID-19 Pandemic Disease. In: Tripathi S. L., Dhir K., Ghai D., Patil S.
(eds) Health Informatics and Technological Solutions for Coronavirus
(COVID-19). (CRC Press, 2021). https://doi.org/10.1201/9781003161066.
[2] Coronavirus History: https://www.webmd.com/lung/coronavirus-history.
[3] COVID-19: A History of Coronavirus: https://www.labmanager.com/lab-
health-and-safety/covid-19-a-history-of-coronavirus-22021.
[4] Kamta N. M. and Chinmay C. A Novel Approach Toward Enhancing the
Quality of Life in Smart Cities using Clouds and IoT-Based Technologies.
In: Digital Twin Technologies and Smart Cities, Internet of Things
(Technology, Communications and Computing), pp. 19–35. (Springer,
2019). https://doi.org/10.1007/978-3-030-18732-3_2.
[5] Symptoms of COVID-19. https://www.cdc.gov/coronavirus/2019-ncov/
symptoms-testing/symptoms.html.
[6] Sheikhzadeh E., Eissa S., Ismail A. and Zourob M. Diagnostic Techniques
for COVID-19 and New Developments. Talanta, 220, 121392, 2020. https://
doi.org/10.1016/j.talanta.2020.121392.
[7] Alpdagtas S., Ilhan E., Uysal E., Sengor M., Ustundag C. B. and Gunduz O.
Evaluation of Current Diagnostic Methods for COVID-19, APL
Bioengineering, 4, 041506, 2020. https://doi.org/10.1063/5.0021554.
[8] Information for Clinicians on Investigational Therapeutics for Patients with
COVID-19. https://www.cdc.gov/coronavirus/2019-ncov/hcp/therapeutic-
options.html.
[9] Lalit G., Emeka C., Nasser N., Chinmay C. and Garg G. Anonymity
Preserving IoT-based COVID-19 and Other Infectious Disease Contact
Tracing Model, IEEE Access, 8, 159402–159414, 2020. 10.1109/
ACCESS.2020.3020513, ISSN: 2169-3536.
[10] Archived: WHO Timeline – COVID-19. https://www.who.int/news/item/27-
04-2020-who-timeline—covid-19.
[11] Timeline of the Corona virus: https://www.thinkglobalhealth.org/article/
updated-timeline-coronavirus.
[12] Akhbarifar S., Javadi H. H. S., Rahmani A. M. et al. A Secure Remote
Health Monitoring Model for Early Disease Diagnosis in Cloud-Based IoT
Environment, Personal and Ubiquitous Computing, 2020. https://doi.org/10.
1007/s00779-020-01475-3.
[13] Cinay D., Brian L. D and Chinmay C. Generative Design Methodology for
Internet of Medical Things (IoMT)-based Wearable Biomedical Devices,
International Congress on Human-Computer Interaction, Optimization and
Robotic Applications (HORA’21), 1–4, 2021, 10.1109/HORA52670.
2021.9461370.
[14] Amit K., Chinmay C. and Wilson J. Intelligent Healthcare Data Segregation
using Fog Computing with Internet of Things and Machine Learning,
International Journal of Engineering Systems Modelling and Simulation, 12
(2/3), 2021, 10.1504/IJESMS.2021.10036745.
[15] Amit K., Chinmay C., Wilson J., Kishor A., Chakraborty C. and Jeberson W.
A Novel Fog Computing Approach for Minimization of Latency in
Healthcare Using Machine Learning, International Journal of Interactive
Multimedia and Artificial Intelligence, 1–11, 2020. http://dx.doi.org/10.
9781/ijimai.2020.12.004.
[16] Parimala Devi M., Raja G. B., Gowrishankar V. and Sathya T. IoMT-Based
Smart Diagnostic/Therapeutic Kit for Pandemic Patients. In: Chakraborty C.,
Banerjee A., Garg L., Rodrigues J. J. P. C. (eds) Internet of Medical Things
for Smart Healthcare. Studies in Big Data, vol 80. (Springer, Singapore,
2020). https://doi.org/10.1007/978-981-15-8097-0_6.
[17] Boopathi Raja G., Sathya T., Ragavi P. and Parimaladevi M. Remote Based
Wireless Framework for Diagnosing Covid-19 using Internet of Things,
International Journal of Advanced Research in Computer and
Communication Engineering, 10(5), 80–83, 2021. https://doi.org/10.17148/
IJARCCE.2021.10514.
[18] Otoom M., Otoum N., Alzubaidi M. A., Etoom Y. and Banihani R. An IoT-
based Framework for Early Identification and Monitoring of COVID-19
Cases, Biomedical Signal Processing and Control, 62, 102149, 2020. https://
doi.org/10.1016/j.bspc.2020.102149.
[19] Chinmay C. and Arij N. A. Intelligent Internet of Things and Advanced
Machine Learning Techniques for COVID-19, EAI Endorsed Transactions
on Pervasive Health and Technology, 1–14, 2021. http://dx.doi.org/10.4108/
eai.28-1-2021.168505.
[20] Muhammad L. J., Ebrahem A. A., Sani S. U., Abdulkadir A., Chinmay C.
and Mohammed I. A. Supervised Machine Learning Models for Prediction
of COVID-19 Infection using Epidemiology Dataset, SN Computer Science,
2(11), 1–13, 2021. https://doi.org/10.1007/s42979-020-00394-7.
[21] Sukanya R. and Chinmay C. Panic Buying Situation During COVID-19
Global Pandemic, Journal of Information Technology Management, 13(2),
231–44, 2021, DOI:10.22059/jitm.2021.80625.
[22] Adafruit IO website: https://io.adafruit.com.
[23] If This Then That (IFTTT) website: https://ifttt.com/.
Chapter 4
Assessment analysis of COVID-19 on the global
economics and trades
Hemanta Kumar Bhuyan1 and Chinmay Chakraborty2
Abstract
The worldwide epidemic known as COVID-19 is being labeled a worldwide pan-
demic on the earlier of 2020, the globe has still not recovered. Cases rapidly spread
from China; leading world governments to take extensive infection control mea-
sures to several cases and restrict the virus’s global transmission rate. However,
these controls have undermined the contemporary world economy’s main sup-
porting pillars of global commerce and collaboration. In light of the epidemic’s
context, this chapter critically evaluates the inventory of the pandemic’s bad and
positive influences in different sections. This advocates for a complete overhaul of
the global economic development paradigm based on a linear economy system that
leverages profiteering and energy-guzzling industrial processes.
Due to the worldwide breakout of the pandemic COVID-19, the world’s
political, social, economic, religious, and financial systems have all been thrown
into total disarray. As of April 2020, an estimated 4.7 million individuals have been
tested, and the illness has resulted in a confirmed infection count of around
2.7 million individuals, with 182,740 deaths attributed to the virus. More than 80
nations have forced companies to shut, locked borders to nations in transition,
quarantined their people, and shuttered schools for around 1.5 billion school-age
children. A total global economic collapse is inevitable because of the world’s
largest economies of different countries. The global financial markets have been
battered, and tax income sources have crashed into a bottomless pit even more
worrisome. Infection is substantially affecting global economic growth. It is
anticipated that if the present growth rate continues, the virus might outpace world
economic growth by almost 2.0% per month. If the global economic slump is deep
and extensive, global commerce might decline from 13% to 32%. It will be years
before the full impact of the outbreak is revealed. It looks into the correlation
between COVID-19 and the development of the national economy and the stock
1
Department of Information Technology, Vignan’s Foundation for Science, Technology, & Research,
Guntur, India
2
Department of Electronics and Communication Engineering, BIT Mesra, Ranchi, India
market in order to prove how well the COVID-19 economic growth prediction is
linked with the gross domestic product. The utilization of publicly accessible data,
as found on Yahoo Finance, the International Monetary Fund, John Hopkins
COVID-19 map, and regression models was used to carry out the goal of this study.
COVID-19 is used to measure the economic effect, and the stock market serves as a
proxy for economic variability to test whether or not the forecast is reliable. In the
aim that the model can make predictions about two quarters out, it is supposed to
provide explanations for changes in the quarter ahead. This study will aid those
with government-level decision-making, business-stage strategic thinking, and
capital-market investment to better comprehend the current state of affairs and use
the model for forecasting.
Keywords: COVID-19; Financial systems; Regression model; Gross

domestic products; Product revenue; World economy growth
4.1 Introduction
The World Health Organization (WHO) proclaimed the new coronavirus (COVID-
19) a pandemic, the world came to attention to a severe issue [1,2]. The virus
quickly spread worldwide, first to Japan, South Korea, Europe, and the United
States. In preparation for the official pandemic proclamation, early this year, sub-
stantial economic signs showed that the globe was leaning in the direction of an
unparalleled watershed in human history if not of our time [3]. Many specialists
within the professional cadres of various industries have made warnings of a global
financial disaster. Expert commentary had predicted an escalating disease crisis
but, by all accounts, the growing chaos brought by the pandemic was more than
what was expected. Based on the number of deceased individuals, the virus has
spread over 800,000 [4]. As a result, livelihoods have been interrupted for millions
of people. At the time of writing, a global recession is looming [5]. Countries
worldwide undertook strict steps to combat the epidemic while also trying to
minimize the number of cases. These included obligatory national lockdowns and
border closures.
The pillars on which these tactics have destroyed contemporary economies
rest. The financial consequences of this epidemic are still being calculated as of
now. Data is in constant motion, government policies often shift, and the virus that
kills its victims spreads from country to country, interrupting manufacturing,
introducing supply networks to dysfunction, and destabilizing financial markets
[6]. Overall, the increasing data indicates that we are at a time when the earth’s
sustainability has to be reconsidered. At the same time, though, inevitable acci-
dental environmental and ecological consequences have resulted from govern-
mental policies. Going ahead, it is now possible to imagine a significant shift to
human biological and physical activity on Earth [7]. But according to [5], noted by
these authors, our investment in global integration and economic growth is no
Assessment analysis of COVID-19 on the global economics and trades 87
longer feasible for stable development. A possible answer to the quest for
numerous economic, environmental, and financial goals is implementing the
financial system.
On the one hand, these packages concentrate on accelerating the existing eco-
nomic status quo. On the other hand, they are dedicated to building a more sustain-
able low-carbon economy. This chapter urges for the recalibration and rethink of the
current global economic growth model, which is predicated on a linear economy
system and uses industrial methods that need excessive use of natural resources and
the pursuit of profit before the environment. Both future resilience and competi-
tiveness are made possible by this plan and the belief that future unintentional that
emerged during the corona virus calamity may be effectively utilized or frozen to
supply openings for both long-term strength and the future.
4.2 Backgrounds
A problem statement cannot be developed when an effective literature review is not

completed, and a literature review illustrates a paradox [8,9]. Additionally, doing a
literature search is critical in pinpointing numerous research questions. A search
and evaluation of the current literature is used to get around this problem. Adopting
the previous conceptual framework was utilized to build empirical work that pre-
sents research derives. Based on the framework, the present research employed
theory to undertake an in-depth review.
Given that it utilizes the critical literature review (CLR) methodology, the
chapter explores the issue under research, critically assessing, synthesizing, and
articulating pertinent materials to help provide new theoretical frameworks and
viewpoints from several disciplines [10]. The problem CLR has when it comes to
subjectivity is amplified by Grant and Booth [11]. They contend that because of the
emphasis on stringent literature selection in systematic literature review (SLR), an
SLR might help to improve the issue. While the study does focus on a particular
part of the pandemic, it looks at the effects of COVID-19 on the global economy
and ecosystems and the benefits of circular economy (CE) initiatives. To achieve
this aim, the author believes that a CLR method is the best choice. It enables for a
diverse variety of theoretical viewpoints and opinions from other sources to be
included [10].
In light of this, the chapter in this collection includes archived data of
newspaper stories, government policy documents, experts’ interviews, expert
opinions, and reports that were pertinent to COVID-19 and the notion of CE.
After reviewing the search results, both theoretical and empirical models were
developed and then utilized to build the CLR. A specific impact analysis was
carried out for the study, using the I ¼ P A T model, where I is impact
defined through population size (P), per capita affluence (A), and technology (T)
maintaining and consuming unit-wise.
It was then explored whether the role of CE as a vehicle for a productive
variance of COVID-19 can continue to exist in the post-COVID-19 world. These
exploratory activities were then completed with observable reflections based on

cumulative data of CEs after COVID-19.
4.3 Social impacts on finance

The least amount of time for an epidemic causes stress on the healthcare system,
and hardship on the economy is few months. In other words, the long-held view
that any sort of infectious illness epidemic is a significant source of localized
economic troubles is still correct [12–14]. History is replete with instances of
endemics, plagues, and pandemics, all of which represent significant outbreaks. In
many cases, this increased the likelihood of states collapsing, healthcare facilities
swamped, social unrest, and creating an economic dislocation that affected other
sectors. Near the beginning months of the COVID-19 pandemic, it has become
apparent that all-natural, incidental, or purposeful risks to global health and the
international economy now represent an unquantifiable risk.
Pandemics have been occurring at a higher frequency during the last 100 years,
according to [15]. Some sources suggest a petite with numerous records of epi-
demics of infectious diseases, which comprise the Black Death [16], the Justinian
plague [17], the Spanish flu [18], HIV/AIDS [19], SARS [20], Ebola [21], and
dengue [22]. This outbreak’s potency fluctuates from outbreak to outbreak.
Because of this, their impact on the economy differs in various retroactive eva-
luations [11]. As an example, in 2013–14, while the Ebola virus was ravaging West
Africa, it cost the countries involved $53 billion [23].
Today’s environment is vastly different from that of previous epidemics. It
is not feasible to make direct parallels between the present the socioeconomic
crisis of COVID-19 and earlier global pandemics, since several aspects have
changed in the last few decades [23] with distinguishing characteristics. Yet,
historical lessons from prior epidemics remain valuable, even though the
modern world is quite different from what was prevalent in the past [24]. Thus,
the CEs strategy in Figure 4.1 shows different historical consequences of
COVID-19.
These include reasons such as the fact that the globe is increasingly inter-
connected because of globalization and technology improvements, which means
that the worldwide pandemic known as COVID-19 is a serious concern [25–29].
Science, health, and engineering have all seen dramatic improvements throughout
the years. Air travel was hindered during the previous pandemics because of the
small number of passengers. There is no longer the case since travel has expanded
significantly worldwide. Due to low-interest rates, there is a surplus of goods, but
demand and supply are balanced at record lows [23]. Even more critically, the
nations experiencing a major impact from the present epidemic are not only lower
middle income nations but also the global manufacturing and supply-chain pow-
erhouse nations. The optimization model can use to analyze for the supply chain as
[30,31]. Based on the previous issues, the evaluation of the effects of COVID-19 is
mentioned in the next section.
Gray literature
Impact of COVID-19 Circular

CE builds a change
(positive and economy (CE)
driver
negative impact) and COVID-19
CE strategy recommendation
after COVID-19
Figure 4.1 The circular economy for COVID-19
4.4 Framework for the international financial system,

bionetworks, and maintainability on pandemic
4.4.1 Assessment strategy constructions to fight COVID-19
The various nations have implemented different methods and policies to address
COVID-19 at different epidemic stages when resources were at a premium. Even
though 65% of global production and exports are done in nations such as China, the
United States, Korea, Japan, France, Italy, and the United Kingdom, these nations
were most hurt by COVID-19 [32]. Due to the inadequacy of hospitals to respond
to situations requiring medical equipment, there has been a focus on acquiring tools
such as personal protective equipment (PPE) and ventilators [33]. We have got to
design frameworks that maximize their utilization alongside bed areas when it
comes to ventilators. Additionally, many businesses have also been negatively
impacted, including mining businesses that have been adversely impacted by
changes in process and international commodity worth. The features of data can be
considered [34–36].
COVID-19 causes panic purchasing because of national and individual degrees
of uncertainty, as detailed in future sections. The shortcomings of the traditional
supply-chain paradigm have been laid bare in each of these cases. They were
concerned about whether the framework would deal with crucial goods when they
were necessary. Following are the worldwide finance repercussions of COVID-19,
along with further discussion on the situation: global production, export, and import
volume.
4.4.2 Macro-finance impacts

One difficulty that the healthcare business faces is that best practices, such as the
JIT macroeconomic framework used in the USA (e.g. stockpile critical medical
equipment), do not financially incentivize the accumulation of vital medical

equipment. This proved to be expensive since governments such as the United
Kingdom, India, and the United States had to implement extreme measures to
secure their supply chain to let companies like Ford and Dyson compete in the
market for ventilator design and manufacturing [37]. Due to the high costs and the
critical scarcity of ventilators, the Defense Manufacturing Act was invoked to get
automakers to concentrate on the production of ventilators. Because of the con-
tinuous shortage of N95 masks, pharmacy industry of different country joined the
worldwide market and look for lower cost equipment [38]. The state of scarcity was
that the United States was accused of “piracy” by other nations in Asia’s medical
equipment supply stocks.
Given the previous information, it may be surprised to learn that China and the
EU waived or reduced their duties on raw materials and PPE, respectively. Both
France and Germany implemented comparable blinkered policies, and the EU
enforced limitations on the exporting of PPEs, placing a lot of nations that had
previously depended on the exporting of PPEs in jeopardy [39]. Logistics problems
and the misallocation of important equipment may cause an increase in the trans-
parency and traceability of emerging technologies like blockchain, RFID, and
Internet of Things (IoT). Technology alone will never solve our future problems,
which need cooperation and scenario planning to supplement it. To help states
avoid incurring federal competition, the EU put a procurement framework for
member states to reduce competition amongst themselves.
On the other hand, in the United States, where state governments had com-
plained that federal involvement with orders had the potential to stifle competition,
a ventilator exchange program was established. Realization creates openings for
manufacturing localization, which comes with the repercussions of developments
in ecological and communal maintainability [32]. It is shown in the instance of
N95 masks, which skyrocketed in popularity enough that both private and com-
mercial flights employed air freight services to provide them instead of using
standard shipping containers [40].
To sum up the findings presented in upcoming parts, there was a notable decrease
in emissions associated with conventional shipping, but at the same time, a rise in the
use of air shipping because of customers’ and suppliers’ desperation and hurry. In any
case, many nations worldwide have had to reevaluate their international prices as a
consequence of COVID-19 epidemic reality. This is because a pandemic in the early
1990s, COVID-19, and several eastern European and Mediterranean nations have all
led to a greater appreciation of national interests and protectionism. The graph in
Figure 4.2 shows the worldwide sell overseas for each nation, compared to China’s
share of the equivalent sell overseas (x-axis and y-axis). The ideal chart for each
product has a huge circle that is located in the upper right corner.
4.4.3 Econometric effects: consumer preferences

There has been disconnecting between people’s material desires and the physical
world for quite some time. The widespread use of COVID-19 has compounded the
Revealed economics log scale
10 20 40
Global market share
Figure 4.2 Global product export from China to other countries as [43]
increased focus on the societal consequences of individual behaviors. In many

nations, the behavior of consumers was first panicky with significant purchases of
food and sanitary supplies. Overall, consumer attitude is also shifting, both pri-
vately and in the marketplace. It has caused people to reevaluate their spending
habits and requirements and prioritize the necessities more. Tech obsolescence is
affecting both the linear economy model and the business-as-usual model. It pre-
dicts mobile phones with a normal lifespan of 2 years (1 year in the United States)
depending on manufacturing and renovating facilities restrained by finance
shutdowns.
Cottage industry production may solve crucial equipment costs, which are all
too costly, rare, high-skilled, and material-intensive. In contrast, however, medical
technology, which might greatly profit from gathering manufacturing and the
customaries of high-tech tool, is saddled with the problem of patents. To assist cope
with health issues like COVID-19, the problem of 3D printer in-frustration in Italy
led to proposals for open pharmacy facilities.
In some countries, factories are running out of facemask inventory and have
to ration their production and distribution. Interestingly, Livingston et al. [41]
described that the handmade facemask business has sprung up to serve the more
significant population. Still, it has also become important in ensuring the supply
of resources and aiding an evacuation route if a lockdown is in place. The res-
urrection of traditional small house-built-up methods of manufacturing tools and
simple but needed goods like facemasks might fundamentally alter the global
manufacturing environment for decades to come, likely lowering levels of con-
sumerism. Due to the high likelihood, a recession will result in short-term views,
which will cause companies to cancel the long-term product. It is primarily for
companies in the automotive and aerospace industries that previously endured
recessions.
4.4.4 Nonpositive impacts of COVID-19

From a contraction of gross domestic product (GDP) in several nations to envir-
onmental and social difficulties, ranging from the very bottom of society to the
upper classes, the negative impacts have been diverse. Socioeconomic activities
effectively ground to stop the progress of various international event aspects. The
entertainment venues were all cancelled, and unemployment claims went up in the
millions.
COVID-19 is a severe pandemic that has caused immense human suffering,
resulted in a catastrophic health, and placed tremendous strain on international
economies [42,43]. It has significantly harmed the global economy, and govern-
ments, businesses, and people have scrambled to deal with it. Figure 4.2 identifies
several global market shares with a revealed economic scale during pandemic
COVID-19, as mentioned in Table 4.1.
As a result, the COVID-19 pandemic has shown the world’s economic para-
digm to be utterly vulnerable to unanticipated disruptions and catastrophes. It has
highlighted the vulnerabilities of centralized global supply and production chains
networks and instability in global economies. This has directly impacted employ-
ment, and as a result, has created an increased danger of food poverty for millions
owing to the locking down of borders and the imposition of border restrictions.
Some government interventions have brought to a leveling of the COVID-19 data
(as mentioned in Table 4.2). As at the time of writing, new healthcare crises are
continually occurring in various world regions.
Despite predicting a modest global economic growth rate of 3.3% in 2020, the
IMF now says that the global economy would contract by 6.3% in the same period.
The potential financial catastrophe this would cause is equivalent to a worldwide
financial shock like the enormous despair and worse, which has already exceeded
the severity of the international economic crisis of 2009 as in Tables 4.3 and 4.4.
With caution, a detrimental impact is projected for emerging and developing eco-
nomics, while recessions are seen to occur in China and India by the end of 2020.
Table 4.1 Different global products shared in a different country as Figure 4.2
S. Global market share Revealed economics scale Colors

no.
1 Fabricated metals Kazakhstan ferroalloys Blue
2 Machinery equipment Bulgaria, Egypt, Hungary, Morocco, Gray
Romania
3 Textiles, footwear Jordan sports clothing, Turkey Green
4 Chemicals and pharmacy pro- Morocco fertilizers Light red
ducts
5 Motor vehicles and transport Hungary, Poland, Romania Yellow
6 Vegetable products Uzbekistan frozen vegetables Deep red
7 Other manufacturing Lithuania, Poland, Romania Light
green
Table 4.2 Month-wise new COVID-19 cases in 10 most affected countries (as
k ¼ 1,000 values)
Feb 2020 Mar 2020 Apr 2020 May 2020 Jun 2020 Jul 2020 Aug 2020
India 0 0.1 1 2 6 23 70
USA 0 0.1 30 31 20 60 70
Brazil 0 0.1 1 5 20 40 50
Mexico 0 0.1 1 5 8 5 4
Colombia 0 0.1 1 2 2 3 6
Argentina 0 0.1 1 2 2 3 3
Peru 0 0.1 1 2 4 3 3
South 0 0.1 1 2 2 3 10
Africa
Iran 0 0.1 1 2 2 3 4
Russia 0 0.1 1 2 2 3 4
Table 4.3 Comparison finance effects between COVID-19 lockdown 2020 and
global financial crisis 2009
Economics item COVID-19 lockdown 2009 global financial crisis

2020
Global economic growth forecast 3.0 0.1
Advanced economics 6.0 3.2
Emerging market and developing 1.0 2.8
economics
Major economies in recession
1. United States 6.0 2.5
2. Euro Area 7.5 4.8
3. Japan 5.0 5.4
4. China 0.1 9.2
5. India 2.0 8.2
Table 4.4 Economic loss over 2020 and 2021
2018 2019 2020 2021

Expected GDP growth 97 100 104 107
Loss of GDP growth 97 100 97 102
The collective GDP loss due to COVID-19 might be over $9 trillion over the
following year.
The current rise in worldwide poverty might be the first since 1998. Up to
49 million people might be forced into severe poverty because of the effect
COVID-19 is projected to have on sub-Saharan Africa. Using the most effective

policy tools, the finance department predicts the growth rate of poverty, inequality,
exclusion, and discrimination over the medium and long run. Global security
structures used as a kind of automatic stabilizers can significantly lessen the
occurrence of poverty and safeguard employees for several years to come.
In Table 4.2, all values are multiplied by 1,000 to make the number of people
affected COVID-19.
4.4.5 Impact of international commercial trading

COVID-19 is expected to contribute to reducing the world economy.
Fragmentation and geographic dispersion inherently plague supply systems. While
it has made organizations more complex and interconnected, globalization has also
made them more susceptible to upheavals. According to an examination conducted
by the United States, most businesses have accounted for commotions in their
trading. This has caused a supply-chain emergency that originated from the short-
sightedness and inflexibility of various supply chains at different layers around the
world and the limited supply diversity that companies use. These disruptions would
affect the loss of production for enterprises in exporting nations and the unavail-
ability of raw materials for enterprises in importing nations. In the end, this will
result in a “momentary manufacturing desert” in which a nation, area, or city’s
production decreases considerably, resulting in a geographic restriction in which
everything except basics, such as food items and prescription pharmaceuticals, may
be obtained. Due to China’s rise to prominence in global supply chains and the
economy, this has resulted. Global commerce is expected to shrink by around 32%
as a result of COVID-19.
A clear example of this is China’s decreased exports, which have led to the
worldwide decrease in economic activity. Trade has now contracted for the second
time since the mid-1980s. Based on various pandemic waves and hazard levels, the
picture illustrates how a pandemic affects global supply networks.
4.4.6 COVID-19’s effect on the aviation industry

As a result of the massive movement and aircraft limitations in place, the trans-
portation industry has borne the brunt of COVID-19.
Aviation, for example, has seen flight cancellations and prohibitions, result-
ing in fewer flights and significant revenue losses. The expenses of operating and
maintaining an airport do not decrease as traffic demands fall, making it difficult
for airports to lower expenses since running and maintaining an airport stays
the same. The full-year 2020 maintains two scenarios for economic aspects:
Scenario 1 (recovery in the latter part of May) and Scenario 2 (start again in the
third quarter). It anticipated with Scenario 1 that the number of airline seats
would be reduced between 39% and 56%, as well as the number of passengers
between 872 and 1,303 million, failing disgusting functioning incomes of
about $153–$231 billion. It is operating with Scenario 2 and predicts that there
would be a loss of between 49% and 72% of airline seats: 1,124 to 1 with a
110
100
90
Index on crisis months
80
70
60
50
For SARS (2003) NAAR
40 Square for avian flu (2005) SA
For SARS (2003) AP
30
× For SARS (2003) CDM
20 + For avian flu (2013) APAR
Dimond for MERS flu (2015) SK
10
–4 –2 0 2 4 6 8 10 12
Crisis on before and after months
Figure 4.3 Effect of past disease outbreaks on aviation
corresponding loss in revenue of $198–$273 billion. They stated that the pro-
jected outcomes are dependent on the length and magnitude of the pandemic and
on the number of funds put toward managing the crisis, the degree of consumer
trust for air travel, the state of the global economy, and the speed of the country’s
economic recovery [44].
When losses occur in the aviation sector, it is essential to contextualize the
situation, and various other comparisons have also been made about the airline
business. Foreign passengers are expected to decrease from 44% to 80% between
2019 and 2020. Airports Council International predicted that air passenger traffic
would decline by two-fifths and cost $76 billion in 2020 if airport revenues
remain constant. IATA anticipated $113 billion in mislaid income and a 48%
decrease in domestic and international flights. Predicting an epidemic scenario is
mentioned in Figure 4.3. The history of prior illness outbreaks reveals the degree
of their influence on aviation. The 8% drop in RPKs and $6 billion in incomes for
Asia/Pacific airlines have all been down to COVID-19’s influence. While the 6-
month recovery road for SARS seems long, it offers an understanding of global
markets.
However, the aviation industry also has an obligation for several economic
aspects, such as fuel economy and waste utilization, and not just focus on income
loss due to COVID-19. This is rather unfortunate news, as per economical air
travel, since it will likely result in significant job losses. However, in the eyes of
environmentalists, these projections are hopeful, and they should motivate the
airline sector to examine greener, more sustainable methods.
4.4.7 Significant collision on the travel sector

From the results of COVID-19, it is expected that the impact on aviation would
lead to an additional effect on the tourist business, which today is significantly
reliant on air travel. It estimates that international tourism revenue is projected to
drop by $80 billion (a 22% decrease) in 2020, which corresponds to a reduction of
67 million international visitors. Current scenario modeling revealed that the
numbers of foreign visitors might range from 58% to 78% less in the next months,
although the prognosis remains very unpredictable. If the travel restrictions con-
tinue, as many as 120 million direct jobs in the tourist industry would be placed at
danger. From this data, it can be deduced that for the foreseeable future, COVID-
19 has impacted the worldwide travel industry worse when compared to other
international tourist trends over the previous 50 years, resulting in a recent 10-year
stretch of uninterrupted expansion coming to an end. International visitor numbers
are expected to plummet by over 60% this year, lowering the contribution of
tourism to global GDP. Table 4.5 illustrates the impacts of COVID-19 on tourism
in of 2020, concerning a 15% increase in foreign visitors arriving during the first
3 months of the year. COVID-19 has a positive impact on sustainable development
objectives.
UN member states in 2015 agreed to set 17 Sustainable Development Goals
(SDGs) to enhance the standard of living for people and the natural environment by
2030. The SDGs were constructed on two substantial assumptions: globalization
and economic development. The fact that COVID-19 has already hindered this
assumption has been covered in the preceding sections. Coverage provided by
COVID-19 is drawing attention to the fact that the SDGs presently exist cannot
cope with epidemics. Progress was at a standstill before COVID-19, since the
SDGs had not been fully established.
Here we have covered some of the benefits of COVID-19. Despite the vast
array of problems caused, COVID-19 has resulted in some good improvements in
behavior and attitudes. The following topics are on the rise because of the flu, and
they help illustrate the increasing pace incorporate functions that have decreased
productive operations. This has caused considerable decreases in exhaust emissions
from autos, other forms of fuel burning. Earth Observatory satellites demonstrate
significant decreases in air pollution across China and important European towns as
indicated in Figure 4.4. Air pollution in China was reduced by 20%–30%, with
similar reductions in the United States and India, while in Rome, Milan, and
Madrid the air pollution concentration fell by 45%–54%, while Paris had a
Table 4.5 The impact of COVID-19 on tourism in quarter 1 of 2020
World Europe Asia and the Pacific Americas Africa Middle East
January 2 6 2 0 4 6
February 9 6 37 3 4 5
March 57 60 64 46 44 41
30%
Spanish flu World war II Global
Great First financial
depression oil shock crisis
20%
Second
oil shock
10%
0%
–10%
–20%
1900 1910 1920 1930 1940 1950 1960 1970 1980 1990 2000 2010 2020
Figure 4.4 Yearly rate of change with effecting energy demand highlighted
tremendous drop of 54%. Air pollution caused by NO2 dramatically decreased in

significant cities in the United Kingdom. Note that in some regions of China,
pollution mitigation measures are inadequate because of the harmful environmental
impacts; the reduction in human activities cannot wholly compensate because of
adverse alterations to the environment.
The preceding possibilities emphasize the dangers of our current lifestyles and
strong dependence on fossil fuel-based transportation systems. This kind of pollu-
tion created a plethora of respiratory illnesses, including respiratory disease, cor-
onary heart disease, lung cancer, and asthma, among others. Air pollution has
shown to be a substantial environmental hazard to the general health and well-being
of the population. According to research conducted in the United Kingdom,
approximately between 28,000 and 36,000 fatalities per year is associated with
enduring disclosure to air pollution. It has been stated that the decrease in air pol-
lution during the lockdown time in conjunction with the subsequent developments
in air excellence had saved more lives than would have been spared, otherwise due
to COVID-19 in China.
Air pollution has decreased; however, the reduction in ambient noise has been
minimal. A significant number of Europeans, about 20% of the population, suffer
long-term exposure to harmful noise levels. About 48,000 new occurrences of
ischemic heart disease each year and approximately 12,000 early deaths are due to
environmental noise pollution. Over 22 million individuals are chronically irritated
while around 6.5 million individuals are impacted by severe restlessness. About
12,500 pupils had noise-induced hearing loss as a result of the planes flying
over them.
According to a recent survey, pandemic COVID-19 enhanced global levels of
physical activity by less traffic congestion and lower noise and pollution. As seen
previously, average, moderate, and inactive exercisers have seen that their exer-
cise program frequencies go up by 88%, 38%, and 156%, correspondingly
improved beach cleanliness. Beaches defend the land from maritime storms and
cyclones and are essential for coastal natural capital assets. They contribute to the
sustainability of coastal communities by providing (such as tourism and leisure as
Table 4.5) valuable services. They are precious since overexploitation must be
controlled. Changes have been noticed because of the COVID-19-induced
measures.
4.4.8 Significant reduction in primary energy usage

Globally, energy consumption decreased by around 3.8% in the first quarter of
2020 compared with 2019, with a substantial impact that can be seen in March [45].
Suppose curtailment measures, such as limited mobility, are maintained over
lengthy periods. In that case, economic recoveries will be delayed, and the energy
demand would decline by up to 6%; therefore removing the last 5 years of energy
required increases. In Figure 4.4, IEA predicts that global energy demand would
plummet to levels not seen in the last 70 years if their predictions come to fruition.
The impact will be 7.4 times greater than the 2008 financial catastrophe.
Alternatively, if COVID-19 is discovered sooner than expected and the economy
experiences a rapid recovery, the drop in energy might be limited to around
4% [45].
Although 8% less coal was produced in the first quarter of 2019 compared to
the same period last year, overall demand was higher owing to China’s heavy
reliance on coal, cheaper gas, renewable energy growth, and warm weather con-
ditions. Oil consumption fell by over 5% in the first quarter as a result of trans-
portation and aviation-related constraints accounting for over 60% of global oil
consumption [45]. Some road and air transportation industries activity was about
50% and 60% below the 2019 average. With complete lockdown limitations in
place, global power consumption decreased by nearly 20%. This corresponded with
a spillover impact on the energy mix. Similarly, renewable energy sources’ pro-
portion in the overall energy supply expanded due to increasing capacity and more
priority dispatch, given that their outputs are generally unhampered by consumer
demand [45].
4.4.9 Record decrease in CO2 emissions

COVID-19’s massive drop in energy consumption was the primary cause of the
substantial decrease in worldwide greenhouse gas emissions. Although no such
event has occurred, CO2 emissions are forecast to continue falling at a pace never
seen before. The 8% emissions reduction will likely lead to the most considerable
emission decrease ever recorded (Figure 4.5). That rate of decline is also expected
to be greater than the sum of the five deepest recessions that the United States has
experienced since the 1950s [45]. Because of this, there will be an increase in
emissions after the economic downturn has ended, unless interventions are based
on greener and more robust energy infrastructure [45].
35
30
25
20
Growth
15
10
0
1900 1920 1940 1960 1980 2000 2020
Global energy-related emissions
Figure 4.5 Global energy-related emissions provide a sense of scale [45]
4.4.10 Rise in digitalization

One aspect of the decline in air pollution after the COVID-19 introduction directly
resulted from the reduction in economic activity caused by the lockdown. While
this effect is temporary, though, there is a difference in the structure of the global
economy, as demonstrated by the improvement in air quality after the introduction
of COVID-19. The improvements are not the result of appropriate climate change
policies enacted by governments, and as such, should not be seen to represent a
victory in this area. More crucially, economies will need to recover, and emissions
will rise. To emphasize our thesis, we carried out a decomposition study of four
worldwide air pollution drivers (accelerators or retardants) with the findings pre-
sented in Tables 4.6 and 4.7.
For example, although global CO2 emission increased by 32% from 1995 to
2009, economic activity and emission factor accelerated the increase as shown in
Figure 4.5. Still, the economic structure, emission intensity, and fuel mix each had
a role in slowing down the increase. Although economic activity with emission
factor boosted emissions, the addition of driving variables, such as economic
structure, emission intensity, and fuel mix, mitigated this upward impact. Reducing
flying and driving in response to COVID-19 helped reduce emissions by around
8%. However, zero-emissions are unattainable due to these actions alone. It means
emissions reduction is only possible when an ideal balance is found among the
following four drivers: the economy, policy, technology, and innovation.
Typically, following a recession, such as the global economic crisis of 2008,
the environment experiences an increase in pollution. However, the substantial and
persistent reduction necessary to establish a low-carbon economy will not be
caused by societal reactions. This is visible since we have interconnected supply
networks, multidimensional manufacturing methods, and nonlinear consumer
Table 4.6 Drivers of CO2 emissions for analysis of pollutions
S. no. 1995 1997 1999 2001 2003 2005 2007 2009

1 CO2 emission change
0 10 15 18 13 2 7 10
2 Economic activity
0 5 2 1 10 28 48 50
3 Economics structure
0 5 8 7 11 7 10 10
4 Emission intensity
0 10 15 17 12 0 6 10
5 Fuel mix
0 0 1 1 1 2 3 4
6 Emission factor
0 0 0 1 1 1 2 2
Table 4.7 Drivers of NOx emissions for analysis of pollutions
S. no. 1995 1997 1999 2001 2003 2005 2007 2009

1 NOx emission change
0 5 12 10 5 4 20 30
2 Economic activity
0 11 19 17 27 43 57 34
3 Economics structure
0 0 5 8 2 0 2 5
4 Emission intensity
0 10 7 3 8 25 32 12
5 Fossil fuel dependency
0 0 0.5 0 0.1 0.1 0 0.2
6 Emission factor
0 10 18 16 20 24 40 42
behaviors. While this might enhance the number of environment-friendly alter-

natives available to customers, it may also cause a rise in vehicle ownership and
hence an increase in carbon emissions. Suppose sustainability principles inspired
by the concepts of low-carbon CE are to be expanded. In that case, proper methods
to maintain the right balance in the ecological life that the world can endure are
imperative. Next, the possible use of the CE in combatting such pandemics as
COVID-19 is addressed.
4.5 The role of circular economy

The conventional linear financial system has long been the foundation of the
industrial economy. The unrestricted use of natural resources, regardless of their
long-term sustainability, puts a strain on the planet’s capacity to produce such
resources. Let us take the manufacturing industry example; around 90% of raw
materials are considered trash before leaving the production facility, yet in manu-
facturing, 80% of things made have been disposed of during the first 6 months of
their existence. Similarly, Hoornweg and Bhada-Tata [46] have found that an
annual quantity of 1.3 billion tonnes of solid waste with a total price tag of $205.4
billion. They estimate that the total quantity of this waste could grow to 2.2 billion
tonnes by 2025, which would increase the total price tag to $375.5 billion. Thus,
worldwide demand for resources is expected to increase by 2-fold by the year 2050.
Under these circumstances, it maximizes the ability to decouple financial devel-
opment from raw materials and reduce waste and building capital has intensified.
These three ideas are key in developing a comprehensive environmental sys-
tem design: create goods and materials out of waste, utilize goods and materials,
and regenerate natural systems. CE is focused on (i) I helping companies maximize
the ecological benefits of their manufacturing and product recovery activities; (ii)
bringing about a harmonious ecological partnership between corporations, con-
sumers, and governments; and (iii) moving to a manufactured goods worth that
takes into consideration the environmental impact from inception to end-of-life.
Now is the perfect opportunity to study how the CE principles may be used in
the actual world when the global economy rebounds. The pandemic has made it
abundantly clear financial system, which is falling short on solving environmental
and social problems. Humans have only intensified their interconnections and
interdependencies in the current global pandemic, with their natural environment,
economy, and social systems. This chapter will briefly explore the several possible
uses of CE as a tool in terms of mitigating climate change, helping construct well
financial system, and fostering socially equitable and inclusive societies.
4.5.1 The circular economy for slowing the onset of climate

collapse
The abnormal finding from the shutdown is that while the limitations to movement
and other leisure activities were in place, emissions were at 92% of the starting
figure. Worrying climate change mitigation measures have often been framed as
“prohibiting the beautiful things of life,” yet, as shown, merely cutting off 8% of
these “good things” provides an actual decrease in greenhouse gas emissions. Thus,
reducing emissions from power plants has a cost of around $3,200–$5,400 per
tonne of carbon dioxide in the United States, for example, as reported by the
Rhodium Group [47]. It states that, if an 8% CO2 emission decrease occurs, it
would release an equivalent of about 172 billion tonnes of CO2 instead of 187
billion tonnes of CO2. This implies that a completely new strategy is necessary in
the effort to tackle the climate challenge.
Therefore, as a result, a more robust low-carbon CE model must be widely
adopted in order to combat the findings of researchers who believe climate
breakdown, rather than COVID-19, will be the largest danger to world health. A
considerable reduction in greenhouse gas emissions cannot be achieved with
renewable energy just but with CE methods. Reduce emissions while promoting
climate change mitigation by putting pressure on product life cycle management,

like (i) material recirculation (reducing the production of primary materials while
increasing the use of recycled materials); (ii) with a link to using more recycled
materials while decreasing primary material production; (iii) product material
efficiency (optimizing production processes, utilizing recycled components, and
designing products with fewer materials). Credible remedies to the bulk of the
structural vulnerabilities highlighted by COVID-19 are offered by comprehensively
adopting these principles. Green and CE investment and climate-resilient infra-
structure go hand in hand, playing the twin function of providing new jobs while
also boosting environmental and economic advantages. It is hoped that by adopting
a CE, countries could create more robust economies.
It found that if we use short-term remedies to address subsequent shocks, it
will be greater than the capacity supposed to address them. It means that we have to
come up with long-term risk mitigation and budgetary strategies designed to
replace short-term profit-oriented thinking. When it comes to food goods, locally
sourced things have an advantage since they do not need the rest of the supply chain
to operate. On the other hand, the whole supply chain may be a better alternative
for other things. As COVID-19 continues to indicate, specific individual cycles
have the inappropriate scale level, and this means that a reexamination of cycle
capacity might be seen as an invitation to do so.
The world is seeing a growing realization of interconnectivity among ecolo-
gical life due to the consequences of COVID-19. This supports the use of the ideas
of resilience thinking in enhancing social innovation. It will aim to benefit the
population’s overall well-being rather than only increase the competition and the
market size.
4.5.2 Social finance system

On the other hand, advanced economies have mainly concentrated on helping
families keep their purchasing power. Many emerging nations have also used
similar tactics by including significant increases in social welfare expenditures in
their containment strategies. Nonetheless, such prevention measures also bring
attention to how inequitable the global economy is, since their deployment in more
developed nations would have a devastating effect on poorer nations and people.
Second, in developing countries, working from home is limited due to poor infra-
structure and healthcare access. Such a course of action, which governments have
used over the short term, is thus inadequate at remedying long-term inequalities and
social injustices.
This goal may be reached via CE strategies, which identify resources and
resources/wastes as wastes and then turn them into something useful to decrease
pollution and help meet social inclusion objectives. Several businesses are adopting
this concept. To avoid potential looping and neediness, this closes all loops and
returns sustenance to the hungry. The development of a country’s digital and
materials innovation can be expedited by increasing investment in the CE. Even
more so, it believes that the CE can promote socially equitable and costs that will
be involved in applying various end-of-life treatments and automation to these

processes.
Furthermore, as CE offers workarounds for all these issues, it has the capacity
to give rise to local employment and help “reindustrialize” an area by using labor
instead of energy, materials instead of labor, and local businesses instead of cen-
tralized factories. However, some people would rather see more automation rather
than less. However, this is a political and economic issue rather than a scientific or
technical one in practice.
4.5.3 Hurdles to CE for context of COVID-19

CE’s primary purpose is to generate three victories in three different categories:
social, economic, and environmental. Constraints do, however, remain at every
stage of the transition from the linear economy to the CE. Within these actors and
interactions within the actors, the obstacles reside. Some of the nontrivial factors
identified by Korhonen et al. [48] in their COVID-19 report include (i) limitations
impressed by recycling or remanufacturing (as with combustion in material and
energy); (ii) with material and energy footprints; (iii) with organizational culture
and management methods, which may use recycled materials and energy; (iv)
consumer and organizational inertia (as with business models that utilize recycled
materials and energy, as well as management and culture); (v) fragile industrial
ecosystems (as with cooperation between the organization, region, and subregions
along with pubic authorities); and (vi) the need of standard agreement on what all
the Rs. Manufacturing or supply chain challenges such as unclarity in product
history through time, difficulties in data exchange between product endpoints, and
high start-up costs have also been seen in other countries.
To ground the paradox of COVID-19, the researchers also had to design a
system that would provide a one-time chance to reexamine some of these obstacles.
Still, CE strategy proponents applauded some sharing economy models as out-
standing examples. Ride-sharing modes might help to boost demand in transpor-
tation. Use as pay performance may have a reduction in total bookings by Airbnb of
416% occurs with every doubling of new COVID-19 instances.
As long as greenhouse gas emissions are reduced, this will need significant
investment by developing and implementing new, low-carbon technologies and
infrastructure. Now that COVID-19 has resulted in a worldwide economic crisis, it
is expected that the chances for large low-carbon investments from the business
sector would be less than prior to the slump. This perspective may be used to both
the corporate and governmental sectors as shown by Naidoo and Fisher [5]. Also, as
COVID-19 concludes, one way to accelerate progress in achieving CE is by pro-
viding local, regional, and national governments with decisive financial and legal
as well as a positive environment. By implementing this, the community will have
increased capacity to respond to future pandemics, and it also helps integrate with
several current resource efficiency roadmaps. Several detection phases of COVID-
19 are mentioned in [49]. The classification data and privacy of data can be used for
COVID-19 record as [50,51].
4.6 Chances financial support after COVID-19

Local industry has been encouraged for job development, which has inspired a
behavioral shift in customers and necessitated more variety and circular supply
chains. There is no longer an issue of whether we should rebuild, but how. As a
result, it is just as crucial for governments to decide whether or not they will form a
new route toward socioeconomic progress as it is for local companies and their
customers to collaborate to facilitate the transition to a CE. The pandemic effects of
COVID-19 are already well documented. The various monetary policies used by
various countries to mitigate these short-term implications are presented in pre-
vious parts of this study.
Nevertheless, it has the prospect of fostering a desirable technical and beha-
vioral shift across many countries in the long run. To learn more about pandemics,
reference COVID-19, which deals with pandemics, not all of them, and medical
accessory production, when product development and manufacture occur in the
same location.
For the enormous demand, panic purchasing and the purposeful stockpiling of
crucial medical items for profit are all blamed on the COVID-19 incident. We will
advocate for creative and environment-friendly alternatives to the shortages, such
as reuse and recycling, since reputable entities like the United States centers for
Disease Control and Prevention have already made these recommendations.
However, medical accessories like PPE are created and manufactured using non-
compliant procedures, which means that they are hard to repair and reuse without
causing substantial deterioration in their performance. Design solutions to close
resource loops, such as the McDonough and Braungart model, are helpful in
accomplishing some of the required competencies.
One other rising piece of evidence favoring CE is being embraced without
considering it in the current influenza outbreak: a few different companies could
easily repurpose their manufacturing floors to produce new items to meet the
medical accessory shortage. Notable examples include, but are not limited to,
AMTICO (flooring manufacturing) reconfigured their production lines;
BARBOUR (a clothing company) redeployed their manufacturing resources.
4.6.1 Several solutions to manage hospital medical and

general waste
Because of its toxic, caustic, infectious, corrosive, potentially radioactive, and to
character, it is common for HCI wastes to invoke fears about security. So, health-
care waste generated distant from traditional municipal garbage has to be managed
according to severe regulatory models. On pre-COVID-19, it is stated that the
startling amount of HCI waste reaches a large amount of annual waste product, and
several studies have examined possible solutions for this problem at the national
level. However, in this new era of the pandemic, the other one of the related pro-
blems is the worrying trend among consumers that places emphasis on personal
cleanliness at the price of the environmental effect caused by the plastic packaging
used in food delivery and grocery shopping. We have hardly given attention to the
disposal of these nonbiodegradable plastics, yet most of these items are created
from them. Reaction in the environmental community has arisen due to the treat-
ment of these types of waste, resulting in reasonable apprehension in several sec-
tors. Even frustratingly, there are not many things that can be done in the present.
All that can be done is to devise appropriate waste management policies for these
potentially dangerous materials. When confronted with the complexities of typical
HCI waste management procedures, the effort is overwhelming. Another critical
step in implementing waste reduction and recycling strategies is reducing the
physicochemical complexity of HCI waste. These goals correlate with the CE
mission, which stresses waste reduction above all else.
In a study conducted by Wong et al. [52], the researchers discovered that
different materials were found in hospital wastes. Medical device manufacturers
may take steps to address the COVID-19 threat by increasing the amount of bio-
products. In fact, even with use of bioplastics and recyclable materials, there will be
medical devices and supplies that cannot be made from them. On the other hand,
there is also a lot of resistance to CE in the healthcare business (HCI). However,
environmental gains may be outweighed by health and safety risks and rules that
need for strict compliance. On the other hand, though, CE is beginning to gain
traction in HCI, helping hospital supply chains run more efficiently while reducing
operational costs while at the same time doing good with good life.
For the most part, the use of finance system is related to the movement of
materials and waste evaluation. Classifying 80% of hospital garbage as general
garbage, along with the remaining 20% as infectious trash, has shown that in recent
years, hospitals have almost doubled their general garbage generation since the
1990s. There are two kinds of HCI waste: incineration is a popular way to deal with
both of them. Large quantities are not problematic for incineration since it creates
such toxic products. Thus, lowering the amount of nonhazardous waste by using
green buying practices or lowering waste input in the manufacturing and supply
chain is molded by adopting such activities. While this might be of use in the short
term, we must take a long-term approach in tackling this problem, adopting CE.
This will enable an eventual shift to environment-friendly HCI, beginning with the
life cycle evaluation of medical products and working our way up to the concept of
reusable medical equipment. Voudrias [53] provides several CE solutions for
hospital waste management, as outlined by Kane et al. [54]. The proportion of
infectious trash increases because hospitals are doing everything they can to keep
the epidemic under control. However, it is still possible to realize the potential of
recycling large amounts of general trash using thermal, microwave, or biochemical
sterilizing methods. It should go into all aspects of building and the built environ-
ment to embrace resource efficiency.
Based on the financial section, COVID-19 has called attention to the failings of
present practices in ecology, especially regarding affordable housing and inflexible
building stocks. In some instances, people who lived in inadequate housing and tiny
dwellings with inefficient energy consumption become infected with the virus. The
situation is most apparent in impoverished nations where people cannot access
proper sanitary facilities, hindering them from adhering to safe practices that are
essential to fighting the spread of disease. Despite these concerns, a rising worry
over the industry’s resource-wasting character, coupled with the increasingly pre-
valent belief that something must be done, make a compelling argument for
reconsidering it. As a result, the CE will be positioned to propose viable solutions
to these issues.
Occupancy requirements may provide behavioral issues, and CE may assist to
balance them. Research has shown that humans spend around 90% of their time
inside. It is also probable that office and commercial space renovations need
adapting them to varying demands, such as change in occupant density, social
distance, ventilation, and the like. When there is not enough air, healthcare staff
and patients at temporary hospitals might get infected. The measure’s influence on
energy consumption must be assessed because of social distancing strategies, which
might need lower population density but higher ventilation rates. Therefore, despite
improved energy recovery being very relevant for the built environment, reducing
mechanical ventilation needs for a lesser number of people places extra require-
ments on building energy consumption. Some academics have advocated that
buildings should utilize 100% fresh outside air instead of recirculation (important
for energy savings). It helps to circulate resources and better utilize resources,
helping to reduce CO2 emissions and provide employment opportunities. In the
case of a warehouse, an ARUP study calculated that planning for steel reuse may
lead to a 6%–27% and 9%–43% savings, resulting in material cost reductions of up
to 25%. Policies that the EU is advocating would make recycled materials (a con-
cept known as material looping) a mandatory component in certain building goods
once safety and functionality have been checked. To help designers and researchers
to make their objects suited for circulation, projects of this kind should encourage
them to use material looping in their entire design approach, beginning at the
product design stage and on through the whole value chain. It has been shown that
by using this material winding process, costs for disposal are reduced and sec-
ondary materials may provide new sources of revenue. Reducing building waste
will assist to cut solid waste by a third in India.
4.6.2 Facilities for CE in communication sector

The transport was one of the most severely affected industries as a result of
COVID-19. A range of CE initiatives will likely be employed in the future in order
to help maintain and strengthen the transport industry. It will assist to reorganize
urban fabric and improve transportation infrastructures that have already been built.
These kinds of cities have a higher amount of infrastructural sprawl, need greater
overall mobility, and more operational vehicle usage, causing a rise in traffic
congestion, resource depletion, and pollution. To a certain extent, “social separa-
tion” impacts many urban residents’ perceptions; however, this shift may result in
behavior change for individuals.
A transport sector plan based on urban freight strategies that also benefit
environmental noise, air quality, and waste management is possible for providing
CE services. Transportation may be informed by using the availability of rich

transport data and AI-enabled sophisticated data-processing technology. Routing
traffic in real time and rules on traffic flow, route planning, dynamic pricing, and
parking spot allocation may all be done in real time using real-time data. Because
these innovations involve CE, it is not surprising that there is still a lack of coor-
dination among them. Because governments usually deal with urban planning, the
government must establish an integrated strategy for mobility to guarantee efficient
resource flows. Innovation and improved utilize of advantages and big data solu-
tions are possible through stakeholder involvement in the transport industry.
Developments in air excellence must be sustained.
One of the beneficial developments owing to the COVID-19-imposed shut-
down was improved air quality. A heightened focus on investment in cleaner modes
of public transit, including sidewalks and bike routes for health benefits, as well as
cities are redesigned to make it more difficult for polluted automobiles to enter
inhabited areas. To improve this much, we must increase the adoption of electric
cars via making more ambitious objectives for the use of electric cars and providing
more charging stations for electric cars as well as actively advocating for the use of
alternative fuels with lower emissions.
Batteries provide a big contribution to the drive to a renewable energy system
and decarbonization of road transportation. It is critical to develop a circular,
responsible, and fair battery value chain to actualize these transformations. This
means determining which of the batteries is the one to be identified. Such batteries
are inexpensive and sustainable and provide a framework that supports the imple-
mentation of renewable energy technologies powered by batteries, which results in
more widespread taking of non-petrol vehicles for pollution free. We acknowledge
that, in the case of an electric car ubiquity, the total number of vehicles will remain
constant, as would the number of materials used in their production, which might
lead to substantial social, environmental, and integrity threats to the whole supply
chain. One way to do this is by the use of the levers as mentioned earlier, which
may assist with overcoming these obstacles and with achieving a sustainable bat-
tery value chain. Removing greenhouse gas emissions during developed, addres-
sing public abuses, providing security situations, and increasing manufacturing,
recycling, and remanufacturing are all aspects of this project.
4.6.3 Use digitalization after COVID-19

Supply chain digitalization is enabled by using disruptive digital technologies
(DDTs). Computing environments are all considered a key step in a company’s
efforts to plan for and deal with disruptive events like COVID-19, which is
expected to emerge globally shortly. Ways to help a company generate value in a
circular supply chain include prolonging the usable lifetime of an item and max-
imizing the use of all assets. Gathering information about the location, condition,
and availability of assets is a source of intelligence. The many possibilities afforded
by cutting these cogs might impact both the product and the business model, which
may lead to innovation and value creation. One wonderful example of big data
analytics in the workplace is the ability to speed up procurement decisions; it is

now utilized to help organizations organize and manage their supply-chain inter-
actions in the cloud. Predictive maintenance is enabled by digitalization, allowing
the addition of manufactured goods and avoiding breakdowns. For that reason, it is
a perfectly suitable delivery vehicle for circular supply chains, which allows the
possibility of closing material loops and improving operations.
Movement toward 5G adoption causes setbacks such as delays in completion
of use cases, a wait for the answers to security, competition, and other wireless
communication regulatory challenges. The development of automation and robots
to combat supply-chain disruptions will be much more important because of
COVID-19. Many organizations are automating their manufacturing processes in
order to save costs. COVID-19 is anticipated to have an effect on the speed with
which regulators provide new rules for 5G. That, in turn, will expedite the adoption
of IoT-enabled monitoring devices.
Despite the advantages of DDTs, a conflict remains between their abilities to
produce quantifiable environmental advantages and the issues they involve,
resulting in rebound effects. It utilizes digital services that must be analyzed for not
aggravating the current problems caused by using and discarding so many resour-
ces. Contingencies arise across supply chains when products must be designed,
procured, distributed, used, disposed of, and recycled in ways that may not account
for contingencies. Incentives, regulations, and public policies to help producers
shift from traditional CE.
Researchers identified seven primary forms of policy intervention to support,
progress, and steer the movement toward a CE. They included both impediments and
regulatory failures that try to fix the market and market action that encourages. The
policies were recommended to include integrating systems such as life cycle engi-
neering education, public communication and information campaigns, and educa-
tional curricula with associated research. It develops to establish platforms for
collaboration and knowledge sharing; also, implementing a variety of initiatives,
incubation of various types of sustainability initiatives, including financial/technical
support, incentive programs. Regulatory models like product registration, and war-
ranty extensions, are also implemented. At the sectoral level, resource productivity
strategies and associated targets are put in place. Also, reductions in VAT or excise
taxes are introduced for manufacturing natural elements.
4.7 Conclusions
The wastefulness of paradigm has been shown in COVID-19. While short-term
plans for coping with the rapid spread of the pandemic are unlikely to be long-term
solutions, the approach might work for a while until something better comes along.
Their findings underscore crucial concerns, such as the vital relationship between
pollution in the environment and the pollution generated by cars and industry.
Unrestricted air travel is clearly implicated in the spread of many viral epidemics,
with industries like tourism and aviation being seriously impacted owing to
decreasing passenger numbers. The long-term consequences for the aviation

industry have proven to hit financially, but resulting in positive environmental
effects, will reshape the sector. Sharing economy or P2P models such as Uber,
Airbnb, and many others have emerged and continue to expand with a new gen-
eration of service providers and workers, although these platforms are very
unstable in the face of worldwide systemic shocks.
A simultaneous increase in demand and supply resulted in an overall decline in
cargo transportation and an increase in airfreight, with transatlantic cost per kilo-
gram tripling overnight. Additionally, there have been jobs lost, income inequality
increased, mass increases in global poverty levels, and industry-wide and supply
chain-wide economic shocks. The capability to work remotely has been explored
and tested in several industries/professions and has shown to provide benefits for
employees with regard to decreased commute. Remote healthcare and telemedicine
are no longer considered impractical due to the success that has been shown in a 4-
month worldwide lockdown. There was a commensurate drop in primary energy
use. Machine learning, IoT, and robots have the ability to increase the efficiencies
of production processes and speed up supplier selection procedures while also
boosting supplier relationship management and logistics.
Hospital JIT procurement has failed to provide big quantities of necessary
medical and emergency supplies at short notice owing to shortcomings in current
healthcare delivery systems. Millions of patients and healthcare personnel died due
to a failure to plan and prepare for PPE with clinical tools adequately. Preventive
measures to manage the increase of the disease necessitated an increase in infec-
tious hospital waste, but proper/advanced sterilization processes using radiation
procedures assist in recycling discarded or retrieved items and PPE.
An ecosystem-based food system resiliency requires the following: (i) an end to
the nutrient loop by using regenerative agriculture, anaerobic digestion facilities, and
organic nutrients; (ii) increased value from and organic nutrients through the use of
anaerobic digestion facilities. International collaboration and mutual interests would
be harmed; long-term planning would be put at risk, and a balance would have to be
struck between outsourcing/importing and local manufacturing/productivity. There is
a likelihood of a realignment of value chains since resource-rich nations may capi-
talize on this global epidemic to drive sustainable prosperity, and a new world order
will develop independent of the influence of major powers in technology.
While most buildings were occupied throughout the lockdown, business and
commercial areas were underused, increasing the amount of ventilation needed in
hospitals. For flexible usage, however, buildings may be redecorated to include
flexible walls. Modular construction methods such as those used in China are
anticipated to become more prevalent in the future. Refurbishment and renovation
will highlight newfound energy with decreased carbon emissions and new
employment being generated as old buildings are given a new lease of life.
In the future, it is recommended that those who use circular thinking put their
knowledge and lessons learned into practice in the broader public’s benefit, not
only for the advancement of corporations with finances. Post-COVID-19 specula-
tions are required to move the world toward extra flexibility.
References
[1] Sohrabi, C., Alsafi, Z., O’Neill, N., et al., World Health Organization
declares global emergency: a review of the 2019 novel coronavirus
(COVID-19). Int. J. Surg. 2020, 76: 71–76.
[2] WHO, Coronavirus Disease (COVID-19) Pandemic. World Health
Organization, 2020. https://www.who.int/emergencies/diseases/novel-cor-
onavirus-2019.
[3] Gopinath, G., Limiting the Economic Fallout of the Coronavirus With Large
Targeted Policies. IMF. March 10, 2020.
[4] JHU, MAPS & TRENDS: New Cases of COVID-19 In World Countries.
Coronavirus Resource Center, Johns Hopkins University (JHU) 2020.
https://coronavirus.jhu.edu/map.html.
[5] Naidoo, R. and Fisher, B., Reset sustainable development goals for a pan-
demic world. Nature 2020, 583(7815): 198–201.
[6] Sarkis, J., Cohen, M.J., Dewick, P. and Schröder, P., A brave new world:
lessons from the COVID-19 pandemic for transitioning to sustainable supply
and production. Resour. Conserv. Recycl. 2020, 159: 104894.
[7] Anderson, R.M., Heesterbeek, H., Klinkenberg, D. and Hollingsworth, T.D.,
How will country-based mitigation measures influence the course of the
COVID-19 epidemic?. Lancet 2020, 395: 931–934.
[8] Piguillem, F. and Shi, L., The optimal COVID-19 quarantine and testing
policies. EIEF Working Papers Series 2004, Einaudi Institute for Economics
and Finance (EIEF), 2020.
[9] Ibn-Mohammed, T., Application of mixed-mode research paradigms to the
building sector: a review and case study towards decarbonising the built and
natural environment. Sustain. Cities Soc. 2017, 35: 692–714.
[10] Snyder, H., Literature review as a research methodology: an overview and
guidelines. J. Bus. Res. 2019, 104: 333–339.
[11] Grant, M.J. and Booth, A., A typology of reviews: an analysis of 14
review types and associated methodologies. Health Inf. Libr. J. 2009, 26:
91–108.
[12] Bloom, D.E. and Cadarette, D., Infectious disease threats in the 21st century:
strengthening the global response. Front. Immunol. 2019, 10: 549.
[13] Bloom, D.E. and Canning, D., Epidemics and economics. In Interactions
between Global Change and Human Health, 2004, Scripta Varia 106.
pp. 304–331. Vatican City: The Pontical Academy of Sciences.
[14] Hotez, P.J., Alvarado, M., Basáñez, M.-G., et al., The global burden of
disease study 2010: interpretation and implications for the neglected tropical
diseases. PLoS Negl. Trop. Dis. 2014, 8: e2865.
[15] Saunders-Hastings, P.R. and Krewski, D., Reviewing the history of pan-
demic influenza: understanding patterns of emergence and transmission.
Pathogens 2016, 5(4): 66.
[16] Horrox, R., The Black Death. Manchester University Press, 2013, pp. 1–364.
[17] Wagner, D.M., Klunk, J., Harbeck, M., et al., Yersinia pestis and the Plague
of Justinian 541–543 AD: a genomic analysis. Lancet Infect. Dis. 2014, 14:
319–326.
[18] Gibbs, M.J., Armstrong, J.S. and Gibbs, A.J., Recombination in the
hemagglutinin gene of the 1918 “Spanish flu”. Science 2001, 293: 1842–
1845.
[19] De Cock, K.M., Jaffe, H.W. and Curran, J.W., The evolving epidemiology of
HIV/AIDS. AIDS 2012, 26: 1205–1213.
[20] Lee, J.-W. and McKibbin, W.J., Estimating the global economic costs of
SARS. Learning from SARS: Preparing for the Next Disease Outbreak:
Workshop Summary. National Academies Press, Washington, DC, 2004,
pp. 92.
[21] Baseler, L., Chertow, D.S., Johnson, K.M., Feldmann, H. and Morens, D.M.,
The pathogenesis of Ebola virus disease. Annu. Rev. Pathol. 2017, 12: 387–418.
[22] Murray, N.E.A., Quam, M.B. and Wilder-Smith, A., Epidemiology of den-
gue: past, present and future prospects. Clin. Epidemiol. 2013, 5: 299–309.
[23] Fernandes, N., Economic effects of coronavirus outbreak (COVID-19) on
the world economy. 2020. Available at SSRN 3557504.
[24] McKee, M. and Stuckler, D., If the world fails to protect the economy,
COVID-19 will damage health not just now but also in the future. Nat. Med.
2020, 26: 640–642.
[25] McKenzie, B., Beyond COVID-19: supply chain resilience holds key to
recovery. Oxford Econ. 2020, 24: 20.
[26] Muhammad, L.J., Ebrahem, A.A., Sani, S.U., Abdulkadir, A., Chinmay, C.
and Mohammed, I.A., Supervised machine learning models for prediction of
COVID-19 infection using epidemiology dataset. SN Comput. Sci. 2021,
2(11): 1–13. https://doi.org/10.1007/s42979-020-00394-7.
[27] Gupta, A.K, Chinmay, C. and Gupta, B, Monitoring of epileptical patients
using cloud-enabled health-IoT system. Trait. Signal, IIETA 2019, 36(5):
425–31. https://doi.org/10.18280/ts.360507.
[28] Sukanya, R. and Chinmay, C., Panic buying situation during COVID-
19 global pandemic. J. Inf. Technol. Manage. 2021, 13(2): 231–44.
DOI:10.22059/jitm.2021.80625.
[29] Chinmay, C. and Abougreen, A.N., Intelligent Internet of Things and
advanced machine learning techniques for COVID-19. EAI Endorsed Trans.
Pervasive Health Technol. 2021, 7: 1–14. http://dx.doi.org/10.4108/eai.28-1-
2021.168505.
[30] Bhuyan, H.K., Kamila, N.K. and Jena, L.D., Pareto-based multi-objective
optimization for classification in data mining. Cluster Comput. 2016, 19(4):
1723–1745.
[31] Bhuyan, H.K., Kamila, N.K. and Pani, S.K., Individual privacy in data
mining using fuzzy optimization. Engineering Optimization, Taylor &
Francis, May 2021. pp. 1–19.
[32] Baldwin, R. and Evenett, S., Covid-19 and Trade Policy: Why Turning
Inward Won’t Work. CEPR Press, London, 2020.
[33] Ranney, M.L., Griffeth, V. and Jha, A.K., Critical supply shortages—the
need for ventilators and personal protective equipment during the Covid-19
pandemic. N. Engl. J. Med. 2020, 382(18): e41.
[34] Bhuyan, H.K. and Kamila, N.K., Privacy preserving sub-feature selection in
distributed data mining. Appl. Soft Compu. Elsevier, November 2015,
volume 36. pp. 552–569,
[35] Bhuyan, H.K. and Kamila, N.K., Privacy preserving Sub-feature Selection
based on fuzzy probabilities. Cluster Comput. 2014, 17(4): 1383–1399.
[36] Bhuyan, H.K., Chakraborty, C., Pani, S.K. and Ravi, V.K., Feature and sub-
feature selection for classification using correlation coefficient and fuzzy
model. IEEE Trans. Eng. Manage. 2021. 10.1109/TEM.2021.3065699.
[37] Iyengar, K., Bahl, S., Vaishya, R. and Vaish, A., Challenges and solutions in
meeting up the urgent requirement of ventilators for COVID-19 patients.
Diabetes Metab. Syndr. 2020, 14(4): 499–501.
[38] Solomon, M.Z., Wynia, M. and Gostin, L.O., Scarcity in the Covid-19
pandemic. Hastings Cent. Rep. 2020, 50: 3.
[39] Bown, C.P., COVID-19: demand spikes, export restrictions, and quality
concerns imperil poor country access to medical supplies. In Trade
Protection in the Age of COVID-19. CEPR Press VoxEU.org eBook, 2020,
volume 31, pp. 1–179.
[40] Browne, M., Sweet, M., Woodburn, A. and Allen, J., Urban freight con-
solidation centres final report 10 Transport Studies Group, University of
Westminster, 2005.
[41] Livingston, E., Desai, A. and Berkwits, M., Sourcing personal protective
equipment during the COVID-19 pandemic. JAMA 2020, 323(19): 1912–1914.
[42] WHO, Coronavirus disease (COVID-19): situation Report – 107, 2020, pp. 1–17.
[43] Javorcik, B., Global supply chains will not be the same in the post-COVID-
19 world. COVID-19 and Trade Policy: Why Turning Inward Won’t Work
111, 2020.
[44] Wang, P., Chen, K., Zhu, S., Wang, P. and Zhang, H., Severe air pollution
events not avoided by reduced anthropogenic activities during COVID-19
outbreak. Resour. Conserv. Recycl. 2020, 158: 104814.
[45] IEA, Global Energy Review, The impacts of the COVID-19 crisis on global
energy demand and CO2 emissions. International Energy Agency. IEA
Publications, 2020, pp. 1–56.
[46] Hoornweg, D. and Bhada-Tata, P., What a waste: a global review of solid
waste management, 2012.
[47] Gates, B., COVID-19 is awful. Climate change could be worse. Climate and
the Coronavirus, GatesNotes, 2020. https://www.gatesnotes.com/Energy/
Climate-and-COVID-19.
[48] Korhonen, J., Honkasalo, A. and Seppälä, J., Circular economy: the concept
and its limitations. Ecol. Econ. 2018, 143: 37–46.
[49] Bhuyan, H.K., Chakraborty, C., Shelke, Y and Pani, S.K., COVID-19 diag-
nosis system by deep learning approaches. Expert Syst. 2021: e12776.
https://doi.org/10.1111/exsy.12776.
[50] Bhuyan, H.K. and Ravi, V., Analysis of sub-feature for classification in data
mining. IEEE Trans. Eng. Manage. 2021: 1–15 (Early Published).
[51] Bhuyan, H.K., Dash, S.K., Roy, S. and Swain, D.K., Privacy preservation
with penalty in decentralized network using multiparty computation. Int. J.
Adv. Comput. Technol. 2012, 4(1): 297–303.
[52] Wong, K.-F.V., Narasimhan, R., Kashyap, R. and Fu, J., Medical waste
characterization. J. Environ. Health 1994: 19–25.
[53] Voudrias, E.A., Healthcare waste management from the point of view of
circular economy. Waste Manage. 2018, 75: 1–2.
[54] Kane, G.M., Bakker, C.A. and Balkenende, A.R., Towards design strategies
for circular medical products. Resour. Conserv. Recycl. 2018, 135: 38–47.
Chapter 5
Early diagnosis and remote monitoring using
cloud-based IoMT for COVID-19
Madhura S. Mulimani1, Shridhar Allagi1,2 and
Rashmi R. Rachh1
Abstract
The entire globe has been battling with deadly coronavirus disease 2019 (COVID-
19) pandemic from the time December 2019. Around 190 million people have been
affected by the virus, and 4 million have lost their lives to it. It has adversely
influenced the socio-economical lives of people in almost all countries across the
world. Hence, it is essential to detect the disease at an early stage and ensure that
the transmission of the virus is curbed, in turn, saving the lives of many other
people. With the advancements and developments in the information technology
field, it is possible to diagnose infectious diseases like the current COVID-19
pandemic at an early stage and give proper treatment to the infected. In addition to
analyzing the disease early, many other approaches are employed to deal with this
deadly disease. In this chapter, various Internets of Things that are being used to
track the patients’ health and provide them the necessary care and treatment even in
remote locations have been discussed. Also, machine learning and deep learning for
early diagnosis and remote monitoring have been discussed. An experimental case
study using COVIDX dataset has been discussed along with the results.
Comprehensive experiments have been carried out with varying computed tomo-
graphy (CT) sizes of CT images and an average accuracy of above 80% has been
achieved. In all, how the use of technology in the medical field proves beneficial
and how it can be leveraged even further to control the spread of the diseases has
been elucidated in this chapter.
Keywords: Internet of Medical Things (IoMT); COVID-19; Early detection;

IoMT devices; Machine learning COVID detection
1
Department of Computer Science, Visvesvaraya Technological University, Belagavi, India
2
Department of Computer Science, KLE Institute of Technology, Hubballi, India
5.1 Introduction
An unusual coronavirus called the very severe acute respiratory syndrome
coronavirus-2 (SARS-CoV-2), initially discovered in China’s Hubei province,
Wuhan, at the end of December 2019, has caused a calamitous event and impacted
entire world economy to collapse adversely, further engendering the societal and
political disruptions. As of July 16, 2021, more than 190 million people worldwide
have been infected with coronavirus disease 2019 (COVID-19), and around four
million have succumbed to it, as per the John Hopkins University and Medicine,
Coronavirus Resource Center, United States [1,2].
Once an individual is exposed to the virus, it may usually take about 5–6 days
to show signs. Individuals with slight symptoms may recoup independently, but
others may suffer from severe symptoms, especially if they already suffer from
other health ailments such as diabetes or heart problems. Animals, too, can get
affected by this virus and can transmit the infection [3].
The pandemic has been rising mainly due to global travels and exploitation of
the environment. Hence, it is crucial to recognize any emergent epidemics and
formulate policies to restrain and thwart the spread of COVID-19 contamination
and save the lives of individuals worldwide, which can be done by detecting the
disease early and ensuring proper treatment for the patients diagnosed with
COVID-19 as well as for the safety of public health [4].
Since there was no vaccine for COVID until recently, the only way to cure or
prevent it was through its early diagnosis. This helps to avoid its further spread
wherein the patients are advised to quarantine or isolate themselves. Most of the
patients with COVID-19 stay asymptomatic. Others demonstrate mild-to-moderate
symptoms with high fever, shortness in breathing, mild cough, and signs alike oxygen
capacity dipping or sometime lung auscultation. These all signs and indications play a
decisive role in ascertaining any infection and execution of further diagnostic tests [5].
Several preventive tools and measures that are being used to prevent the
infection from spreading use personal protective equipment, properly used face-
masks, face-shields, use of gloves, and ventilators [4], washing hands repeatedly
with soap for 25–30 s, continuing social distance, masking mouth and nose with a
disposable tissue paper while sneezing and so on [6]. In addition to all this, to curb
the spread of the disease, there is a need for expeditious and precise testing of
COVID-19 in people, considering the furious spread of the disease.
This major contribution of the chapter includes a comprehensive coverage of
various Internet of Medical Things (IoMT) devices that are being used in the
medical field for diagnosing and remote monitoring of patients with different dis-
eases, procedures carried out during early diagnosis of COVID-19 and remote
monitoring of home isolated or quarantined patients, an experimental case study
that uses machine learning (ML) with deep learning (DL) techniques to distinguish
between the computed tomography (CT) images of normal and COVID-19-infected
patients mainly to identify and diagnose if the patient has been infected with
COVID-19, and various measures that different countries are adopting to thwart the
spread of the virus from affecting more people.
Early diagnosis and remote monitoring using cloud-based IoMT 117
The organization of the chapter is as follows: Section 5.2 discusses about

techniques used for COVID-19 detection. Section 5.3 presents an architecture of
IoMT devices. In Section 5.4, we discuss about various medical devices with
Internet connectivity that are used for identifying COVID-19 symptoms and remote
monitoring. Section 5.5 presents the phases that occur during the identification and
treatment of patients affected with COVID-19. In Section 5.6, we briefly discuss
about how ML and DL techniques can be used or are being used for early detection
of COVID-19. Section 5.7 lists the works of different authors who have used ML
and DL techniques in COVID-19 prediction. In Section 5.8, we provide details
about the experimental case study that was conducted using ML with DL techni-
ques on COVIDX dataset containing numerous CT images of varying sizes.
Section 5.9 lists various preventive measures adopted by different countries for
controlling the spread of the virus. Section 5.10 lists the limitations encountered
when using IoMT devices. Section 5.11 concludes the chapter and gives a brief idea
about how the current work can be enhanced or improved in future scenarios using
recent or emerging developments in the information technology field.
5.2 Detection techniques
People who are infected by the virus and exhibit symptoms such as fever, cough,
body aches, running nose, headache, and sore throat undergo specific tests. The
following two categories of tests are used for COVID-19 detection:
1. Diagnostic tests: Reverse-transcription–polymerase-chain-reaction (RT-PCR) is

a diagnostic test used for detecting an active coronavirus infection. Reverse real-
time PCR assay, a typical molecular-based assay, is a standard diagnostic testing
method widely used to screen COVID-19. Generally, it provides results in 4–6 h.
This test has several drawbacks: time-consuming, reliance on well-equipped
laboratories and trained specialists, manual, expensive, ineffective, laborious,
and complex. Additionally, the number of available test kits is limited in quan-
tity, and to make things worse, domain experts cause further delay in the situa-
tion as the test is manual, laborious, complex, expensive, and ineffective.
2. Antibody tests: Here, tests look for antibodies formed in the body with
response to the disease. The antibodies usually take several days or even few
weeks to grow well and may remain in the blood for a long time even after
recuperation. Hence, these tests cannot be used to detect COVID-19. With
certain limitations in the current investigative techniques and a rapid increase
in number of infected patients, it is a paramount prerequisite that the physicians
use certain alternate automatic screening systems to quickly detect and isolate
patients infected with COVID-19 [5].
The purposes of the two classes of tests overlap in the supervision of the
COVID-19 pandemic. The first class of testing recognizes persons infected with
SARS-CoV-2 during the critical phase of contamination. In contrast, the second
class of testing recognizes individuals within whom antibodies to the virus have
developed, and hence, could be prospective convalescing plasma donors. It can

trace and observe the status of immunity of individuals and assemblages over a
certain period. Hence, to manage COVID-19, it is essential to have an appropriate
identification, active treatment, and future anticipation. This has led to a competi-
tion to establish economical and point-of-contact test kits and well-organized
laboratory practices to confirm SARS-CoV-2 infection [7].
Thorax CT is more reliable, functional, and quicker than RT-PCR to classify
and assess COVID-19 in an epidemic region. Nearly all hospitals carry out CT
image screening. Hence, to detect COVID-19 patients early, thorax CT can be used.
However, a radiology expert and ample time are required to classify COVID-19 on
the basis of thorax CT. Since COVID-19 test results take a complete day to sense
the virus in the human body, it is vital to identify the disease in the initial period
and instantaneously quarantine the infected patients as there are no drugs that are
explicitly existing for COVID-19. As per the government of China, RT-PCR can be
used to diagnose the confirmation of COVID-19. However, the RT-PCR test does
not yield satisfactory results due to its higher false-negative rates. Also, it is time-
consuming because it takes around 4–8 h for the machine to process the samples of
patients. Perhaps, in certain cases, the diseased may not be identified on time and,
hence, may not obtain appropriate treatment. A diseased may be ascribed as
COVID-19 negative due to the false-negative outcome. Hence, in such cases, it is
necessary to analyze the thorax CT images in an automated manner, which can be
time-saving for the specialist medical staff as well as avoid delays in starting the
treatment [3].
Another drawback of RT-PCR is that it needs a laboratory kit that cannot be
easily provided to several countries, especially during crises and epidemics.
Another matter of concern is that this test is not error-free and is prejudiced. In this
test, a skilled laboratory technician extracts the nasal and throat mucosa sample.
Since this method is painful, many people refuse to undergo the test. Also, the RT-
PCR test has a sensitivity of 35–65%, which indicates a diminution in the accuracy
of COVID-19 diagnosis [2].
Since patients with COVID-19 generally remain asymptomatic, diagnosing
COVID-19 early is challenging. It is hard to distinguish between patients having
symptoms and those having various respiratory infections like pneumonia and
severe flu. Since the wet-lab tests for diagnosing COVID-19 are highly priced and
time-consuming, there needs to be some substitute that is noninvasive and fast as
well as a discounted automatic screening system. As a substitute model to detect
and determine COVID-19 infection, a chest CT (CCT) scan can be exploited.
Medical practitioners use imaging modalities such as X-ray and CT as diag-
nostic techniques that do not require any instruments to be introduced in the body to
analyze and cure many diseases. They use these imaging modalities to even
determine COVID-19. However, since the CCT scans are 3D images and contrast
dyes, their performance in diagnosing COVID-19 is significantly improved com-
pared to the simple 2D chest X-ray (CXR). Meanwhile, COVID-19-infected
patients’ CT scans show disparate features with subtle variations and make it quite
challenging to interpret these scans manually. Additionally, the current immense
increase of infected patients has made it more difficult for domain experts to
complete the diagnosis on time. Hence, some computer-aided diagnostic systems
that generally use CT scans are needed to manipulate better and understand the
CCT images. ML approaches have been successfully applied to diagnose lung
diseases. They have also been used to analyze COVID-19 using CCT images with
unpredictable sources and amounts of data [5].
The sensitivity of CT scan images in diagnosing and detecting COVID-19
cases is higher. However, their low-specificity represents that CT scan is extra
precise in COVID-19 cases but less in nonviral pneumonia instances [2]. CCT
scans have been used to detect COVID-19-related lung damage effectively.
However, this diagnostic has some limits.
The CT findings are conventional in a few patients at the beginning of the
infection. Hence, using the CT alone causes it to have a negative prognostic value.
Since the CT scans have a high false-positive rate, it could lead to difficulties in
detecting non-COVID-19 cases. Moreover, if it necessitates for patients to have
manifold CT scans in the course of the disease, the CT scanner rays can harm them
and cause complications.
Hence, the American College of Radiology advises against CT scans as the
first line of diagnosis. Since a CT scan is expensive and there is disease transmis-
sion risk while using the CT scan device, it leads to serious complexities for the
patient and the healthcare systems. Hence, it is advisable to replace a CT scan with
CXR radiography if medical imaging is needed.
Compared to CT scans, X-ray imaging is inexpensive and has widespread
usage. Since the digital image need not be transferred from the contact point to the
examination point, the diagnostic process is accomplished much more swiftly.
Chest radiography is beneficial and quick for the medical triaging of patients. X-ray
imaging expects less occasional and less affluent equipment [2].
5.3 Internet of Medical Things

Internet of Thing (IoT) is an emerging technology in which devices that collect and
share real-time data without human intervention are interconnected and build a
system. Such a system can control various healthcare systems, logistics, traffic
congestion, retails, emergency service, smart cities, etc. The Internet provides an
infrastructure with many benefits such as state-of-the-art connectivity, services, and
systems and, thus, drives past machine-to-machine evolution and is interconnected
with other services. Therefore, it is feasible to introduce automation practically in
almost all fields, and as a result, IoT offers an extensive variety of applications and
solutions [8].
Smart devices, smart sensors, and progressive lightweight communication pro-
tocols have made it conceivable to interconnect medical things and monitor biome-
dical signals, which can further be used to analyze patients’ diseases without any
human involvement. Such medical things combined with intellectual devices are
together labeled as the IoMT. They can transfer data over a system without insisting
on the interaction between humans or between humans and computers [9]. They find
healthcare and medical care applications and are based on communication between
machines through wireless connectivity to transfer data. IoMT finds usage potenti-
ality in various applications such as health monitoring, home medication, fitness
projects, chronic ailments, elderly care, and compliance with treatment. Hence, smart
devices include multiple sensors and medical devices, imaging devices, and diag-
nostics, and they form a primary part of the IoMT, the major benefits of which
include an increase in the standard of living, reduced cost, and increased user’s
knowledge in healthcare systems [8]. Hence, an IoMT platform can be defined as an
intelligent system comprising various elements such as sensors and electronic cir-
cuits, a processing unit, a network device, a storage component, and a visualization
platform. Sensors and the electronic circuits procure biomedical indicators from a
patient that are operated on by the processing unit. A network unit transmits the
biomedical information over a network and stores it in a storage unit that may either
be temporary or permanent. A visualization framework with artificial intelligence
(AI) schemes decides as per the convenience of the physician. The IoMT architecture
is presented in Figure 5.1.
IoMT systems are used in varied areas and are very ubiquitous. They play a
crucial role in the healthcare sector and provide real-time surveillance using
wireless body area networks, AI and the cloud-based remote health testing, remote-
healthcare-monitoring (RHM), and wearable health-monitoring devices. They also
IoT application Data translated to value
Data is processed and

Cloud server managed
Data is forwarded to
Public internet network layer
network
IoT devices APIs
Data is generated Data is collected and sent

for processing
Figure 5.1 Architecture of IoMT

provide real-time, continuous, and remote monitoring of patients. They can make
healthcare more convenient and efficient, especially regarding accuracy, visibility,
accessibility, reliability, continuity, and real-time tracking. For example, hospita-
lized patients suffer massive money on the patients, healthcare centers, and insur-
ance companies. In addition, it is extremely difficult for all those patients who live
in remote locations and cannot easily access either the hospitals or caregiver cen-
ters. In such cases, they may need to travel lengthy distances to pursue healthcare.
However, with the advancement in IoT, they are being used for RHM due to their
characteristic features such as interoperability, communication, and information
exchange. This, in turn, has enabled RHM to provide an enhanced and continuous
monitoring for chronic diseases. Hence, they qualify to be excellent candidates that
can predict, prevent, and monitor any emerging transmissible diseases such as
COVID-19. In the IoMT system, end-user devices such as health monitors, tags, or
cell phones use wearable sensors/devices to gather the remote data for patients.
They then store the collected data in the cloud databases and send the data to the
cloud platform for real-time analysis and application, i.e., for caregivers’ decision-
making and analytics. Hence, the IoMT devices can be leveraged as an early
indication system to control infectious diseases [10].
5.4 IoMT devices for the identification of COVID-19

symptoms and remote monitoring
Millions of people worldwide are using smartwatches and other wearable devices
that detect and gage many physiological parameters like sleep, skin temperature,
and heart rate. Consumer wearable devices continuously measure vital signs, which
can monitor the onset of infectious diseases. For example, consumer smartwatches
are being used to detect coronavirus disease pre-symptomatically. These devices
determine individual baseline parameters of health and detect any substantial
aberrations from the reference physiology readings at the start of infection. Hence,
they are a widely deployed technology. These parameters can be used in a retro-
spective manner to detect COVID-19 early and for real-time health monitoring and
surveillance. For example, fitness trackers acquire the heart rate signals that can be
used well in advance of symptom onset to detect COVID-19 infection.
Wearable devices have sufficient potential to diminish the ill effects of COVID-
19. Since the commencement of the pandemic, millions of people have been infected,
and millions of people have succumbed to this deadly disease. Hence, it is vital to
have an improved infection-tracking system. It can be done using population-scale
technology solutions to identify the infected cases in real-time and track them.
Currently, PCR assays are used to identify active infections, and after illness, they
require 3–4 days for a dependable positive signal. In addition, since PCR tests are
rarely used and most infections become obvious just upon the onset of symptoms, the
existing testing procedures may not detect pre-asymptomatic carriers. Hence, it is
incredibly challenging to implement intrusions at an early stage to reduce transmis-
sion. Supposedly, almost 60% of persons with COVID-19 are asymptomatic and,
thus, enable more viral spread. Nevertheless, to detect COVID-19 early on in real-
time, there is a contingency for accessible and inexpensive methods [11].
5.4.1 Wearables
The fight against the COVID-19 pandemic may be easier when apposite wearable
devices are used. The usage of wearable devices has proved to be an efficient and
effective way to diagnose the disease early during the pandemic as they can detect the
COVID-19 symptoms of the patients. For instance, a wearable IoT device can be used
to determine whether the respiratory symptoms of a patient are normal or not, which, in
turn, can help the patient to witness changes, if any, in their health state and accordingly
choose to have a medical appointment prior to the appearance of any other symptoms.
These technologies are defined as a blend of electronics and anything that can be
worn. They may be described as app-enabled computing technologies that use built-in
Bluetooth wireless technology to receive and process input when worn on the body or
stuck. These intelligent wearables include bands, glasses, and watches and are
designed for various purposes to be used in diverse domains like healthcare, fitness,
and lifestyle. IoT-wearable devices include smart wearable tools such as smart helmet,
smart thermometers, IoT-Q-band, smart glasses, proximity trace, and EasyBand [12].
5.4.1.1 Smart thermometers

These devices are used to record constant measurements of body temperatures and
are accurate, affordable, and user-friendly and can either be worn or stick to the skin
below the apparel. Touch, patch, and radiometric are various forms in which they are
available. These smart thermometers are highly recommended for capturing the body
temperature against the infrared thermometers that may probably spread the virus
more as the patients and healthcare providers are quite close to each other when
checking the temperature. Some smart thermometers can post the body temperature
on a smartphone at any time, including Ran’s Night, iFever, Tempdrop, and iSense.
By means of these devices, the likelihood of detecting novel patients at an initial
phase of the infection can drastically improve in our daily existences.
5.4.1.2 Smart helmet

These wearable devices have a thermal camera and lower human interactions and
are safer to use than the infrared thermometer gun. For example, KC N901, a smart
helmet manufactured by China, detects high body temperature with an accuracy of
96%. The thermal camera on the intelligent helmet detects a high temperature, and
an optical camera within the helmet sends the locale and picture of the person’s
face together with an alarm to the assigned mobile device, which is usually to the
diseased person’s health officer who can take appropriate action. Figure 5.2 illus-
trates the process. To obtain all the places that a suspected person has visited, the
smart helmet can be incorporated with Google Location History.
5.4.1.3 Smart glasses

This is another type of wearable device. These devices have lesser human inter-
actions when compared to thermometer guns. Smart glasses use optical and thermal
Helmet with
optical and
thermal camera
Alarm Processing
Notify for high temperature
Figure 5.2 Smart helmet
cameras, and the built-in face detection technology makes it easy to monitor
crowds, and the tracking procedure once suspicious cases are detected. Moreover,
since Google Location History can capture the places visited by the suspicious
subject, it can effectively empower further actions with more excellent reliability.
An example of such a smart glass is the Rokid that uses infrared sensors and
monitors up to 200 people. Once these devices detect people with elevated tem-
peratures, they make the recorded information available to medical centers or
authorities.
5.4.1.4 Robots
A robot is often “a machine bearing resemblance to a living being in its capability
to move independently.” The evolution in the robotic field has led to the devel-
opment of networked robots within the cloud, and such robots are called the
Internet of Robot Things. These robots can perform various repetitive tasks, make
life easy and simple. Considering the current COVID-19 pandemic situation, dif-
ferent robots such as telerobots, autonomous robots, social robots, and collaborative
robots are being used.
IoT-linked robots can be used in the initial diagnosis of the infection. These
robots remarkably process a patient’s treatments and lower work stress levels for
the healthcare workers. These autonomous robots can help in combating all phases
of COVID-19. For example, autonomous swab test robot can collect throat swabs
samples from patients and help in the process of diagnosis, and as a result, it can
prevent the medical team from getting infected. Vayyar Imaging and Meditemi are
the two companies that have jointly developed a robot to carry out this process,
called the Intelligent Care Robot. It uses a quick scanning of a person within a span
of 1-m without touching any control or interface and captures the breathing signs
and thermal reading to detect the COVID-19 symptoms in just 10 s. Xenex disin-
fection robot is used to sterilize virus and bacteria-contaminated areas like hospitals
and helps in reducing the possibility of infection for the healthcare staff.
5.4.1.5 Drones
A drone is simply an aircraft that is flown with little or no human operation with the
help of a remote monitor. It is an aerial vehicle with no crew onboard and is called
the unmanned or uncrewed aerial vehicle (UAV) and functions using GPS, sensors,
and transmission services. Drones with IoT implementation within them are called
the Internet of Drone Things, and such drones perform varied tasks such as mon-
itoring, searching with delivering. The smart drones are usually operated using a
smartphone and a controller unit and require a least amount of time and energy.
Hence, they can be resourcefully used in various fields such as defense, agribusi-
ness, and medicare. IoT-based drones include a range of drones such as disinfectant
drones, medical drones, thermal imaging drones, multipurpose drones, surveillance
drones, announcement drones. Multipurpose drones that are usually used in the
healthcare domain are now primarily being used to combat the COVID-19
pandemic.
In the initial diagnosis phase of COVID-19, it is imperative to find the infected
people in the crowd. To hasten the process of finding the infected people and areas,
the most common way is to use the UAV, and more specifically, the IoT-based
drones during the pandemic. Since drone technology reduces human communications
and reaches difficult-to-access locations, it is much more useful, especially during the
COVID-19. Combining such drones with virtual reality (VR) as a vesture gadget can
help detect individuals with elevated thermal readings. A Canadian company has
developed the Pandemic Drone application that remotely monitors and detects any
cases of infection. It captures the biological parameters such as temperature, any
sneezing or coughing, and respiratory signs such as heart rate [12].
5.4.1.6 IoT buttons

These are small, programmable, electronic buttons that are usually connected to the
cloud via wireless communication. Each IoT button has an associated code written
on the cloud to perform different repetitive tasks. Hence, pressing a button enables
the code to be executed, and as a result, the intended task is completed. For
instance, Visionstate has produced an IoT button called Wanda QuickTouch. This
IoT button, installed as a cleaning alert system in hospitals, alerts the authorities
regarding the necessary sanitation or public safety concerns [13]. Another IoT
button may enable the patient to indicate a drop in his oxygen level when under-
going a COVID-19 treatment just by pressing a button. Sefucy IoT button, initially
designed to track missing children, is now being used during quarantine for
emergency notifications as a rejoinder to COVID-19 pandemic. If the patient’s
condition, secluded at home gets worse, he can press the IoT button, which will
immediately alert the healthcare provider or the family members [12].
5.4.1.7 WHOOP strap

It is a wearable strap with a photoplethysmography sensor and is worn on the wrist.
The device contains a thermometer, photoplethysmogram, capacitive touch sensor,
gyroscope, and accelerometer. This small strap, which is waterproof and rechargeable,
can be comfortably worn 24 h per day, with each charge lasting up to 5 days. It
collects the metrics such as resting heart rate (RHR), respiratory rate, and heart rate
variability (HRV) and transfers data via wireless communication to the mobile devi-
ces on which the connected WHOOP app runs. From that point onward, data is
conveyed to a WHOOP system that consists of protected cloud-based data storage and
processing server. The system provides physiological data as follows:
● Respiratory rate: The average number of respirations per minute represents the
respiratory rate. The photoplethysmogram present in the strap uses photo-
plethysmography to derive this value every single night for the key sleep
period duration.
● Resting heart rate (RHR): It represents the average values of beats per minute
(bpm) inspected in the previous 5-min duration of the past occurrence of slow-
wave sleep each night.
● Heart rate variability (HRV): It uses the root mean square of successive RR
interval difference method to investigate the slow wave sleep sample during
the last 5–10 min of the previous incident each night. It is expressed in units of
milliseconds.
WHOOP system is usually used to track physiological data automatically.
With additional functionality, it has been able to track COVID-19 symptoms as
well as test outcomes during the ongoing COVID-19 pandemic. Since the primary
stages of the infection may have measurable signs, recognizing SARS-CoV-2
condition before and during the first days of symptoms could help recognize all
those entities who need to self-isolate or quarantine and seek testing [14].
5.4.2 Smartphone applications

They are application-software intended to perform restricted tasks inside a mobile
gadget such as a smartphone. Since there are more than 3 million operational
smartphones in 2021, IoT-based smartphone applications are being efficiently used
in numerous domains like retail, healthcare, and agriculture. Among them, smart-
phone applications such as Stop Corona, nCapp, DetectaChem, Aarogya Setu,
SelfieApp, StayHomeSafe, TraceTogether, HaMagen, Coalition, BeAware
Bahrain, Social Monitoring, and Civitas have been advanced especially for the
healthcare purview and used exclusively in response to COVID-19 [12].
5.5 Early diagnosis of COVID-19 and remote monitoring

procedures
Early detection of COVID-19 among infected individuals and their remote mon-
itoring during home isolation or quarantine treatment can be divided into different
phases as early diagnosis, quarantine time, and post recovery. The procedures
carried out during each stage have been explained in the following:
1. Phase 1: Early diagnosis
The essential step that can be taken to battle the COVID-19 pandemic and
prevent the dispersion of the virus is to diagnose it early. This can be a
considerable help to the healthcare providers as it will enable them to plan

better treatment and arrangements, rescue more lives, and reduce uncleanliness
and infections. As an initial step toward early diagnosis, understanding the
symptoms is most crucial as COVID-19 has wide-ranging symptoms, along
with fever, cough, chills, tiredness, muscle or body aches, headache, etc.
Among all these symptoms, fever or elevated body temperature is one of the
most prevalent COVID-19 signs, and when gauged, it surpasses 38 C or
100.4 F. The detection process can be faster and more efficient when IoT
devices are used as they use sensors within them to capture data that is ana-
lyzed for healthcare providers, patients, and authorities, who will further
detect, control, and eventually end the transmission of the deadly virus that has
caused the pandemic and affected so many lives [12].
The first and foremost task is the collection of real-time symptom data using
a set of vesture sensors on the user body. To detect these symptoms, several
biosensors are available. For instance, various sensors can be used to detect
different symptoms like temperature-based sensors to detect fever, auditory
sensors with acoustic and aerodynamic prototypes to detect cough along with
its classification, locomotion-based and heart rate sensors to sense weariness,
image-based classification to detect sore throat, and lastly, oxygen-based sen-
sors to detect difficulty in breathing. Mobile applications may be used to gather
the travel and contact history in the past 3–4 weeks duration on an ad hoc basis
as such data is also relevant for early diagnosis [15].
Many patients at their homes or hospitals can be monitored with support
from IoT devices. They can collect various vital signs such as blood pressure,
heartbeat and transmit them to the cloud so that the healthcare workers can
analyze the data. This avoids exposing the healthcare workers to the infection.
In one study, it has been demonstrated how IoT was used to detect a fever and
how it automatically uploaded the readings to an Android-based smartphone
and then to a worldwide network via wireless Bluetooth communication.
Consequently, the results were quickly available anywhere, at any time
immediately. Such an IoT system is vital in tackling infectious diseases,
especially for medical practitioners [16].
2. Phase II: Quarantine time
Once an infected person is detected, it is mandatory to isolate them and then
monitor the patient at home or in a hospital. This quarantine is applicable to the
confirmed patients, doubted patients, and diverse areas, cities, or nations. This
is mainly carried out to impede the likely spread from suspected cases or zones
to others. Employing the IoT devices during this period can help alleviate
critical tasks like spread of virus. Hence, patients are monitored skillfully, and
their respiratory signs, heart rate, blood pressure, etc. are controlled [12].
Hence, data is collected from quarantined or isolated users who are in a
healthcare center during the quarantine time. The data consists of both tech-
nical and nontechnical information. While the technical data includes the time-
series data of various symptoms, the nontechnical data includes information
about the travel and contact history in the former 3–4 weeks duration, age,
long-standing diseases, gender, and any other pertinent information like a

family record of ailment [15].
During the COVID-19 pandemic, most IoMT systems have been used as
digital surveillance systems to track and locate the infected individuals. IoMT
has also been used to detect, monitor, and test, to curtail the possibility of
infection or to accelerate the assessment procedures [17]. Though it has steered
to some privacy issues, many citizens have accepted it as they understand that
it is much needed to end the pandemic.
3. Phase III: Post recovery
To restrict the spread of the COVID-19 pandemic, many restrictions like
lockdowns, social distancing had/have been imposed, resulting in an adverse
consequence on many businesses, markets, and the economy.
After many months, the restrictions are being lifted, and countries are slowly
and prudently opening again. During this phase, all and sundry must stay with
extra cautiousness following societal distancing and restrictions on physical
amenities to ensure that the virus will not spread again [12].
Wearable devices such as smartwatches produce much larger datasets that
are almost impossible for humans to analyze. Data mining and computational
approaches based on ML come to the rescue and have been in continuous
demand in such cases. These approaches help analyze the complex relation-
ships between sensor data and the patients’ physiological status, thus leading to
more accurate predictions that were previously unknown. Wearable electronic
devices in healthcare have a great potential to monitor physical as well as
biochemical markers continuously. COVID-19 pandemic has emphasized such
capability in the current wearable devices, and through their continual inno-
vation and development, they could play a significant part in fighting any
future pandemic.
Though it may be easy to collect the physiological data of individuals,
making accurate predictions about the infected individuals using population-
based data and modeling may be inadequate. To decide regarding an indivi-
dual’s health, collecting and processing large volumes of data from every
individual are essential. The cost of low-power silicon electronics is reducing
continuously, and computational power is increasing. This makes it possible to
process the data locally and, thus, decentralize and accelerate the decision-
making process [17].
5.6 Machine learning and deep learning in COVID-19

diagnosis
Clinical testing and image testing, which do not have many transformations, differ
mainly in their procedures. In clinical testing forms, filing procedures take a long
time and are annoying, especially in urgent situations. In contrast to this, image
testing does not have long processes to be followed. In clinical testing, doctors
themselves analyze the reports. This testing is slightly more accurate than image
testing. One needs to rely on the computer, which will intelligently perform tasks
until precise instructions are given. These two types of tests differ mainly in terms
of cost. Image testing is more economical than clinical testing. CT scans can also
be evaluated for other viruses to distinguish them from the coronavirus using dif-
ferent DL-based techniques [3].
ML and DL approaches have established themselves as valuable approaches
that can assess gigantic high-dimensional features of medical images. CT or X-ray
discoveries of COVID-19 sufferers and that of other uncommon diseases like
pneumonia are quite alike. ML and DL approaches can be used to discern between
COVID-19 and other pneumonia conditions robotically. COVID-19 can also be
discerned from other diseases using drug-induced viruses or immune-pneumonitis.
Various DL approaches such as VGG-16, Ensemble, InceptionNetV3, MobileNet
v2, ResNet, Xception, CNN, KNN, and Truncated Inception Net have been used to
evaluate chest images of COVID-19 patients, and their solicitation has produced
hopeful results. This is significant as X-rays are easily acquirable and cost little.
The significant advantage of these methods are that in addition to diagnosing
COVID-19 patients, they can even forecast the austerity of COVID-19 pneumonia
and the threat of fleeting mortality. The number of studies conducted to measure
each of the X-ray and CT images is almost the same, even though the X-rays are
less expensive when compared to the CT images. Some other research studies have
applied procedures on both kinds of images. Several independent studies have
shown that convolution neural networks (CNN)-based methods have classified
COVID-19 patients from further causes of pneumonia or even related ailments with
an accuracy of 99.99%. This means that these approaches can be used as screening
procedures for the primary assessment of COVID-19 cases.
Though both ML and DL approaches can be used for the stated objective, they
vary in some regards. For example, ML can use a minor amount of data from the
users, while DL requires a massive volume of labeled training data for a succinct
denouement. ML methods require the users to accurately define the features.
However, DL methods require high-performance hardware. DL generates features
on its own and, hence, consumes more time to train when compared to ML. ML
divides the tasks into smaller ones and later combines the obtained results into one
conclusion. On the contrary, DL uses end-to-end principles to resolve the problems.
DL is a rapid and efficient method used for the identification and forecast of
various ailments mainly due to its good accuracy rate. Hence, it can be used in
medical science as the most efficient technique. Models are explicitly trained to
categorize the inputs into various classes as per the programmers’ preference. Such
models are used in the medical field for multiple tasks such as detecting heart
problems, diagnosing cancer and tumors using image analysis, and several addi-
tional applications. They are also exploited to discriminate between CT scan ima-
ges of the patients diseased with COVID-19 as positive or negative. In the medical
field, DL method such as CNN efficiently processes CT scan images to the parti-
cular model. Since a CT scan image gives a detailed image of a specific area, it
helps to detect internal defects, tumors, measurements of the parts, injuries, etc.,
and thus, a CT scan proves to be more reliable than RT-PCR method in the case of
COVID-19 and efficiently classifies the images of COVID-19 patients with great
accuracy at a faster rate.
Additionally, during the COVID-19 pandemic, they were able to detect the reti-
cular pattern, ground-glass-opacities, alliance, and crazy tiling pattern, which are the
hallmarks of COVID-19 infection. However, CT scans have side effects, such as
patient exposure to radiation when numerous CT scans are conducted. Diagnosing the
disease early largely increases the chances of treating the disease or the virus in a
better way. Sometimes a few minor symptoms can steer in diagnosing a major illness
like cancer or a minor disease like a simple viral, which can be wrongly predicted [3].
5.7 Related works

Muhammad et al. [18] have developed various supervised learning models using
epidemiology labeled dataset to predict COVID-19 infection and evaluated their
performances. Their results show decision tree, support vector machine, and naı̈ve
Bayes to be the best among all others in terms of accuracy, sensitivity, and speci-
ficity, respectively. Dash et al. [19] have developed a model that predicts the out-
break of COVID-19 and is used to predict the daily-confirmed cases for 90 days
future values, thus, helping to plan and manage the healthcare systems and infra-
structure. Chakraborty and Abougreen [20] have presented a review of the AI and
ML contributions to confront the pandemic disease. Dhawan et al. [21] have pro-
posed a secure technique that uses IoT protocol and steganography to transfer the
stego images to their destination using the secure transmission of hidden images.
Kumar et al. [22] have proposed to classify and audit the coughing sound investi-
gation, AI models, and information assortment strategies through IoT to group the
pulmonary sicknesses. They have tested the feasibility of the model by comparing it
with other models that have used data from a pediatric office and wearable sensors.
Shelke and Chakraborty [23] have presented a review to evaluate the relevance of
augmented reality and virtual reality in spinal navigation.
5.8 Experimental case study

In this section, a case study has been discussed and experimental results are
provided.
5.8.1 Dataset description

COVIDX dataset that is publicly available on GitHub has been used. Our experi-
ment used only a subset of the COVIDX CT dataset due to the limited computation
facility. For the analysis, we have used 100,500 CT slices that span across 4,000
patients. All the CT images are labeled manually with the assistance of medical
professionals, and some are labeled using automatic labeling models. Table 5.1
gives the instances of labeled images used in experimentation.
The dataset is categorized into two variants: A and B. Variant A consists of
COVID confirmed cases certified by medical professionals. The second variant, B,
Table 5.1 Instances of labeled images
Patient id Source Country Finding Verified Slice View Modality

finding selection
CP_0 CNCB China Pneumonia Yes Expert Axial CT
CP_10 CNCB China Pneumonia Yes Automatic Axial CT
....
...
study_0621 MosMedData Russia COVID-19 No Automatic Axial CT
has all image sets of A with additional images confirmed to be diagnosed as

COVID but no validation. These other images are used as a part of the testing
dataset. Figure 5.3(a) and (b) provides the data distribution across normal, pneu-
monia, and COVID.
5.8.2 Methodology
In this segment, we discuss the proposed model for screening patients’ CT scan
images using DL models such as CNN. The model is iteratively trained with nor-
mal and COVID-infected CT images. The images in the dataset are preprocessed to
fit in the considered model. The proposed model consists of the following stages:
(1) Preprocessing and (2) EfficientNet-B0 architecture.
The flow diagram for the proposed model is given in Figure 5.4.
1. Preprocessing:
The image dataset has been captured in real time. The intensity of pixels across
an image varies. For any CNN, it is essential to normalize the pixels across the
Data distribution of variant A

Training Testing
31,020
21,400
18,000
5,000
4,500
3,580
Normal Pneumonia COVID
(a)
Data distribution of variant B
Training Testing
44,520
21,400
18,000
8,500
4,500
3,580
Normal Pneumonia COVID

(b)
Figure 5.3 (a) Data distribution of variant A and (b) data distribution of variant B
image. For this purpose, we have used pixel intensity normalization of all
pixels in the range [0, 1]. This process also helps in the training stage of model
convergence.
For any of the CNNs, images are to be reduced to standard size. Once the
images are reduced, there is a scope of loss of data. Since the cost involved in
EfficientNet is low in terms of memory and latency, we do not have to reduce
the size of the image. Thus, the information in image can be retained.
Figure 5.5(a) and (b) shows the sample images of COVID CT scan and Normal
CT scan, respectively.
COVIDX-dataset
Training dataset Testing dataset
Variant A Variant B
Classification
Preprocessing
Model
Size reduction
convergence Evaluation metrics
Model training
Image augmentation
EfficientNet-B0 architecture
Figure 5.4 Flow diagram for proposed model
2. EfficientNet-B0 architecture:
The EfficientNet-B0 baseline architecture belongs to the family of multi-
objective neural networks. It is further developed for B1–B7 to achieve a
higher accuracy rate. Table 5.2 describes the architecture.
The three parameters, width, depth, and resolution are scaled using compound
coefficient parameters ∅ for the proposed model. The alpha, beta, and gamma scaling
multipliers are used as per (5.1)–(5.3) to compute the depth, width, and resolution.
Depth D ¼ a∅ (5.1)
(a) (b)
Figure 5.5 (a) COVID-19 CT scan and (b) non-COVID-19 CT scan
Table 5.2 Efficient Net baseline network-B0 architecture
Stage-Xi Operator Fî Resolution H îXW î Channels Ci Layers Lî

1 Conv 3 3 224 224 32 1
2 MBConv1, k3 3 112 112 16 1
3 MBConv6, k3 3 112 112 24 2
4 MBConv6, k5 5 56 56 40 2
5 MBConv6, k3 3 28 28 80 3
6 MBConv6, k5 5 14 14 112 3
7 MBConv6, k5 5 14 14 192 4
8 MBConv6, k3 3 77 320 1
9 Conv1 1/Pooling/FC 77 1,280 1
Width W ¼ b∅ (5.2)
Resolution R ¼ g∅ (5.3)
w:r:t:a b g 2; a 1; b 1 and g 1
2 2
The proposed model uses various residual blocks with different configurations
with swish activation functions and inverted residual blocks. Using swish activation
is to minimize the error and increase the accuracy as ReLu produces unexpected
changes. The Swish activation function can be represented as (5.4). Here, it does
not nullify the negative values as ReLu:
Sx ¼ X Sigmod ðX Þ (5.4)
5.8.3 Training
In DL models, large datasets need to be used in the training model because of the
complexity in computations involved. But, in real time, availability of a large
dataset is challenging, and hence, to address this issue, we employ the technique of
data augmentation. We increased the sample size by transformation operations over
Data
augmentation
augmentor
Figure 5.6 Image augmentation operation
Figure 5.7 Augmented image in dataset
images such as rotation, scaling, and flip. Figure 5.6 gives the sample instance of
data augmentation operation.
The image augmentations are performed using the python augmenter package.
The transformation parameter applied for rotation is 0–20 anticlockwise. For
scaling, it is up to 30% zoom. The operations are involved in different probability
orders on samples. Figure 5.7 shows an augmented image in the dataset.
5.8.4 Experimental setup and results

Experiments have been accompanied on an Intel Core i7-5820K CPU with 3.30
TM
GHz, Titan X Pascal with 12 GB, 64 GB Ram, and the TensorFlow with Keras
framework for Python. The training model initiates with a limit learning rate of
0.001 with the Adam-Optimizer for 15 epochs using categorical cross-entropy loss.
The experiment for the proposed model is carried out with varying input image
sizes, and various performance metrics are computed and shown in Table 5.3.
Figure 5.8 gives the ROC curve for the proposed approach.
Table 5.3 Performance metrics for various input image sizes
Input size Accuracy Sensitivity Prediction F1-score

250 250 69.41 60.41 74.15 76.70
300 300 72.81 69.39 78.00 74.12
350 350 75.49 72.59 80.41 78.00
400 400 81.25 85.61 94.01 92.49
450 450 88.55 89.63 93.43 92.17
500 500 (original) 94.74 94.67 98.93 93.55
ROC curve
1.0
0.8
True positive rate
0.6
0.4
0.2
ROC curve (area = 90.51)

0.0
0.0 0.2 0.4 0.6 0.8 1.0
False positive rate
Figure 5.8 ROC curve
5.9 Measures for monitoring and tracking COVID-19

To prevent the transmission of the virus and affecting more people, various digital
monitoring tools are used, such as
1. Social contact tracing: To inhibit the transmission of the virus, many countries
have imposed lockdown rules. However, the lockdown has many downsides
too, such as it interrupts productivity, affects people’s living conditions, and
brings down the economy for the nations. The contact tracing method has been
imposed in a lot of places to control the spread of the virus while simulta-
neously permitting people to live a new normal. This approach is used to
identify potentially infected people based on the study of people’s social
interactions. Thus, the information obtained from the approach can be used to
notify the healthcare workers and people of their risk of infection [4].
2. COVID-19 monitoring using smartphones: Around the world, several
smartphone applications have been developed and deployed. Official gov-
ernments in China, India, and Singapore support some of them. Different
countries have used different approaches to monitor the spread of COVID-
19. Some countries have made it mandatory for citizens to enter their data.
Other countries are automatically tracking the movement of citizens with the
help of authorities who monitor law enforcement. These tracing applications
locate phones using Bluetooth in the vicinity that uses the identical appli-
cation. The Bluetooth signals contain data such as how often individuals
were close to each other, the distance between the individuals. The result of
a person testing positive is uploaded to the app, and depending on how close
other people were to the person, a notification is sent to them who may have
been exposed to the virus.
3. COVID-19 tracking with search queries and news coverage: Any keywords in
Google search queries or any daily news coverage related to COVID-19 are
analyzed using ML to track COVID-19 [12]. Cloud services may either be paid
or free and use different transmission models like Zigbee, Bluetooth, Wi-Fi, for
short-distance transmissions and LTE, Sigfox, etc., for long-distance trans-
missions, which are especially used to transmit data to the cloud and are
included on the IoT. IoT defines the connection between humans and devices
that allow all or a subset of the devices to connect to execute the intended job
exclusive of human interaction [24].
5.10 Limitations of using IoMT devices
One method that can limit the spread of the disease is to use a surveillance system
that can track the infected and exposed individuals and the clinical outcomes.
However, since these surveillance systems have limitations regarding factors such
as spatial resolution, timeliness, and scalability, reporting in these systems has the
tendency to be specific to a nation or a region. They may not have adequate
evidence about maladies at the community or city level. This affects the social
distancing and quarantine measures, and as a result, the disease cannot be con-
tained [16].
IoT and innovative medical devices can monitor and screen patients’ health
conditions anywhere, anytime. As a result, the healthcare systems have improved
tremendously and can provide timely services in various locations. However, due to
the novel COVID-19 pandemic, there has been an unexpected and massive increase
in patients. Hence, it is imperative to constantly monitor the patient’s health con-
ditions to avoid any serious disorder or infection. The IoT and smart medical
devices produce gigantic volumes of sensitive information about patients.
However, the patients do not want their personal and sensitive information to be
disclosed. This raises serious security concerns concerning IoT data [25].
5.11 Conclusion and future scope

COVID-19 has afflicted the lives of many people across the globe. Almost all
countries are taking various measures to control the virus from spreading and to
save peoples’ lives. In this chapter, the advancements in information technology
being leveraged in various ways to diagnose the diseases and monitor the patients
even if they are in remote locations have been described. Wearable devices are
being used to collect the patients’ data that is analyzed using ML and DL techni-
ques to make predictions about the diseases. As a result, it helps to adopt measures
to curb the infection from spreading even further. Additionally, a case study along
with experimental results has also been discussed.
In the current work, ML and DL techniques have been used with available
dataset. However, in real time, since a large amount of data is being continuously
generated on a number of devices, the current work can be extended to deploy
continual machine learning (CML) models in the production environment where the
data is generated, i.e., on the edge devices. CML can retrain the ML model peri-
odically with new data and trends to build a more accurate model with improved
performance, thereby saving the retraining time. In other words, CML can be com-
bined with edge computing to provide autoadaptive models on the edge devices with
some processing occurring at them rather than sending all the data to the cloud and as
a result, data can remain secure as it is processed on the edge devices.
References
[1] Seshadri DR, Davies EV, Harlow ER, et al., Wearable Sensors for COVID-
19: A Call to Action to Harness Our Digital Infrastructure for Remote
Patient Monitoring and Virtual Assessments. Frontiers in Digital Health.
2020; 2: 8.
[2] Ghaderzadeh M and Asadi F, Deep Learning in the Detection and Diagnosis
of COVID-19 Using Radiology Modalities: A Systematic Review. Journal
of Healthcare Engineering. 2021: 1–10.
[3] Shah V, Keniya R, Shridharani A, Punjabi M, Shah J and Mehendale N,
Diagnosis of COVID-19 Using CT Scan Images and Deep Learning
Techniques. Emergency Radiology. 2021; 28: 497–505.
[4] Allam M, Cai S, Ganesh S, et al., COVID-19 Diagnostics, Tools, and
Prevention. Diagnostics. 2020; 10(409): 1–33.
[5] Abbasi WA, Abbas SA, Andleeb S, et al., COVID: An Expert System to
Diagnose COVID-19 and Predict Its Severity Using Chest CT Scans:
Application in Radiology. Informatics in Medicine Unlocked [Internet]. 2021;
23(100540):1–8. Available from: https://doi.org/10.1016/j.imu.2021.100540.
[6] Khanday AMUD, Rabani ST, Khan QR, Rouf N and Mohi Ud Din M,
Machine Learning based Approaches for Detecting COVID-19 Using
Clinical Text Data. International Journal of Information Technology
(Singapore). 2020; 12: 731–9.
[7] Carter LJ, Garner LV, Smoot JW, et al., Assay Techniques and Test
Development for COVID-19 Diagnosis. ACS Central Science IN FOCUS
[Internet]. 2020; 6: 591–605. Available from: https://pubs.acs.org/doi/10.
1021/acscentsci.0c00501.
[8] Bharati S, Podder P, Mondal MRH and Paul PK, Applications and
Challenges of Cloud Integrated IoMT. In: Studies in Systems, Decision and
Control [Internet]. Springer International Publishing; 2021. p. 67–85.
Available from: http://dx.doi.org/10.1007/978-3-030-55833-8_4.
[9] Vishnu S, Jino Ramson SR and Jegan R, Internet of Medical Things (IoMT)
– An Overview. In: ICDCS 2020–2020 5th International Conference on
Devices, Circuits and Systems. 2020. p. 101–4.
[10] Aman AHM, Hassan WH, Sameen SAL, Attarbashi ZS and Alizadeh M, IoMT
amid COVID-19 Pandemic: Application, Architecture, Technology, and
Security. Journal of Network and Computer Applications. 2021; 174: 1–20.
[11] Mishra T, Wang M, Metwally AA, et al., Pre-Symptomatic Detection of
COVID-19 from Smartwatch Data. Nature Biomedical Engineering
[Internet]. 2020; 4: 1208–20. Available from: https://www.nature.com/arti-
cles/s41551-020-00640-6.
[12] Nasajpour M, Pouriyeh S, Parizi RM, Dorodchi M, Valero M and Arabnia HR,
Internet of Things for Current COVID-19 and Future Pandemics: an
Exploratory Study. Journal of Healthcare Informatics Research. 2020; 4(4):
325–64.
[13] Chai PR, Zhang H, Jambaulikar GD, et al., An Internet of Things Buttons to
Measure and Respond to Restroom Cleanliness in a Hospital Setting:
Descriptive Study. Journal of Medical Internet Research. 2019; 21(6): 1–10.
[14] Miller DJ, Capodilupo JV, Lastella M, et al., Analyzing Changes in
Respiratory Rate to Predict the Risk of COVID-19 Infection. PLoS One
[Internet]. 2020; 15(12): 1–10. Available from: http://dx.doi.org/10.1371/
journal.pone.0243693.
[15] Otoom M, Otoum N, Alzubaidi MA and Etoom Y, An IoT-based Framework
for Early Identification and Monitoring of COVID-19 Cases. Biomedical
Signal Processing and Control. 2020; 62(102149): 1–16.
[16] Gong M, Liu L, Sun X, Yang Y, Wang S and Zhu H, Cloud-based System
for Effective Surveillance and Control of COVID-19: Useful Experiences
from Hubei, China. Journal of Medical Internet Research. 2020; 22(4):
1–9.
[17] Ates HC, Yetisen AK, Güder F and Dincer C, Wearable Devices for the
Detection of COVID-19. Nature Electronics. 2021; 4(1): 13–4.
[18] Muhammad LJ, Algehyne EA, Usman SS, Ahmad A, Chakraborty C and
Mohammed IA, Supervised Machine Learning Models for Prediction of
COVID-19 Infection using Epidemiology Dataset. SN Computer Science
[Internet]. 2021; 2(1): 11. Available from: http://link.springer.com/10.1007/
s42979-020-00394-7.
[19] Dash S, Chakraborty C, Giri SK, Pani SK and Frnda J, BIFM: Big-Data
Driven Intelligent Forecasting Model for COVID-19. IEEE Access
[Internet]. 2021; 9: 97505–17. Available from: https://ieeexplore.ieee.org/

document/9474452/.
[20] Chakraborty C and Abougreen A, Intelligent Internet of Things and
Advanced Machine Learning Techniques for COVID-19. EAI Endorsed
Transactions on Pervasive Health and Technology [Internet]. 2021; 21(26):
e1. Available from: http://eudl.eu/doi/10.4108/eai.28-1-2021.168505.
[21] Dhawan S, Chakraborty C, Frnda J, Gupta R, Rana AK and Pani SK, SSII:
Secured and High-Quality Steganography Using Intelligent Hybrid
Optimization Algorithms for IoT. IEEE Access. 2021; 9: 87563–78.
[22] Kumar A, Abhishek K, Chakraborty C and Kryvinska N, Deep Learning and
Internet of Things based Lung Ailment Recognition through Coughing
Spectrograms. IEEE Access. 2016; 4: 1–12.
[23] Shelke Y and Chakraborty C, Augmented Reality and Virtual Reality
Transforming Spinal Imaging Landscape: A Feasibility Study. IEEE
Computer Graphics and Applications [Internet]. 2021; 41(3): 124–38.
Available from: https://ieeexplore.ieee.org/document/9112351/.
[24] Hasan MW, Covid-19 Fever Symptom Detection based on IoT Cloud.
International Journal of Electrical and Computer Engineering. 2021; 11(2):
1823–9.
[25] Akhbarifar S, Javadi HHS, Rahmani AM and Hosseinzadeh M, A Secure
Remote Health Monitoring Model for Early Disease Diagnosis in Cloud-
based IoT Environment. Personal and Ubiquitous Computing. 2020.
Chapter 6
Blockchain technology for secure COVID-19
pandemic data handling
Agbotiname Lucky Imoize1,2, Daisy Osarugue Irabor3,
Peter Anuoluwapo Gbadega4 and Chinmay Chakraborty5
Abstract
In recent times, COVID-19 pandemic data collected via the Internet of Medical
Things–enabled channels can be transmitted seamlessly and processed remotely by
medical practitioners globally. However, the confidentiality and integrity of these
channels pose serious security concerns. The transmission of medical information,
which comprises highly sensitive personal data, must meet stringent security
requirements. In order to achieve this, blockchain has been deployed to enhance the
security, confidentially, trust, and integrity of sensitive data being transmitted over
insecure channels as a result of its cryptographic characteristics. Blockchain is
regarded as a protected and distributed structure of data, which enhances security,
simplifies the tracking, and analyses the stored medical information independently.
Additionally, blockchain allows the transmission and exchange of messages
between two parties in a network configuration, independent of sole trusted
authority. Blockchain technology is verifiably safe against an attacker who mis-
manages the scheme and compromises the central controller. Furthermore, block-
chain technology has been deployed to handle distributed medical data in standard
medical laboratories. The decentralized and immutability features of blockchain
have facilitated the rapid development of the beyond 5G wireless services for
health-related data sharing and processing. However, the adoption of blockchain
technology in medical data handling is still in its infancy, and the need for a
1
Department of Electrical and Electronics Engineering, Faculty of Engineering, University of Lagos,
Lagos, Nigeria
2
Department of Electrical Engineering and Information Technology, Institute of Digital Communication,
Ruhr University Bochum, Bochum, Germany
3
Department of Electrical and Computer Engineering, Tandon School of Engineering, New York
University, New York, USA
4
Discipline of Electrical, Electronic & Computer Engineering, School of Engineering, Howard College
Campus, University of KwaZulu-Natal, Durban, South Africa
5
Department of Electronics and Communication Engineering, Birla Institute of Technology, Mesra,
Ranchi, India
rigorous study in this domain cannot be overemphasized. Toward this end, this
chapter first highlights how blockchain technology has developed in recent times
and how it is applied in the medical data-handling space. Further to this, the chapter
examines the potential benefits, key challenges, and prospects in blockchain tech-
nology. Additionally, the chapter highlights research efforts in blockchain in
healthcare, especially as it relates to the dreaded COVID-19 pandemic. The chapter
also discusses the application of blockchain technology in healthcare data-handling
practices that can build trust with automated tracking of integrity and responsible
credential verification. Finally, practical COVID-19 pandemic data were analyzed
and presented to motivate the chapter further.
Keywords: Blockchain technology; Security; COVID-19 pandemic; Medical

information; Distributed data handling
6.1 Introduction
The present inconsistencies of channels through which COVID-19 pandemic data are
transmitted and collected have given rise to the increasing need for better secured, yet
more reliable, data accessible to health professionals and policy makers [1]. Globally,
these data help manage the spread of COVID-19. Through the Internet of Medical
Things (IoMT), they can be received and processed for health and environmental
benefits. The sensitivity of these data demands the highest level of channel con-
fidentiality and integrity, which are currently inadequate. In addition, the health data
transmission and processing systems designed to improve treatment outcomes may
suffer centralization and licensure that invariably leads to failure at one or more
points in the transmission or collection process [2]. In response, blockchain tech-
nology addresses this by improving the reliability of health logistics, supporting early
detection of an outbreak, and protecting users’ privacy during the pandemic [3].
Blockchain is a system of distributed databases that work using stable decen-
tralized servers and a computer network [4]. Blockchain technology manages a
record of chronological data transactions that are distributed among available nodes
in a verifiable, node-synchronized, decentralized, cost-effective, and timely manner
[5–7]. As the pandemic prevails even with control measures being developed, the
global healthcare system is consistently strained, and characteristic data have
become a precious resource to curb the spread of the virus. We also live in a
technology-driven world that is transforming rapidly. It is valuable to collect and
record digital activities like sensitive data through smart devices and digital mon-
itoring tools. Individuals give these devices access to manage their personal and
business data directly or indirectly, and even though the devices appear reliable,
data may not be handled appropriately. This gives rise to the misuse of sensitive
data that other digital or examining systems collect the data of users and their
network. In most cases, they use these data dishonorably [8]. Although this may be
a common occurrence within social Internet channels, several sensitive data have
Blockchain technology for secure COVID-19 pandemic data handling 143
still been lost, broken, or diverted within healthcare transmission or processing

systems due to mishandling data.
Blockchain technology consolidates data security such as locations visited by
people who have been infected; medications and test kits from manufacturer to users;
records of remote patient–doctor consultation; credential verification of healthcare
specialists; and the source of malfunctioned test kits [9]. Unarguably, this technology
handles data with a high level of confidentiality, transparency, and improved system
integrity. We shall soon see that these may be optimized for even more efficiency
and better security that may accommodate more data and challenging situations.
With this, blockchain technology may streamline the operations of handling medical
information [10]. Blockchain can be private, public, or consortium [11]. In a private
blockchain that may be preferred for handling sensitive data due to its high
throughput and low-energy consumption, the network is controlled by a specific
provider who is trustworthy and manages preselected participants connected to the
Internet. While the public blockchain operates in contrast, the consortium combines
both operations over a network. While healthcare systems may choose any of these
blockchain types, there are specific requirements and cases, like handling the
COVID-19 pandemic data that may demand the operation of a particular blockchain,
according to the size of participants, the strength of the network, and available
resources. A summary of the key benefits of the three categories of blockchain
technology in relation to medical information management is presented in Figure 6.1.
Diagnosis time
xBlockchain prevents a delay in diagnosis of diseases
Effectiveness
xBlockchain helps to improve the effectiveness of medical interventions
Workforce limitations
xBlockchain resolves healthcare challenges and minimizes patient hospital visits
Increased
throughput
xBlockchain helps clinical specialist serve more patients in good time
Convenience
xBlockchain helps experts to conveniently monitor and manage the care of patient
Revenue
xBlockchain provides several revenue opportunities for healthcare facilities
Figure 6.1 Key benefits of blockchain in the management of medical information

After Bitcoin and Ethereum, the applications developed as the third blockchain
technology have been instrumental to healthcare, supply chain, and Internet of
Things (IoT) [8]. In Section 6.6, an explanation will be provided on how blockchain
can be used to handle healthcare data across the network to build trust. For many
years, the IoT has been revolutionizing the healthcare industry to improve medical
information records, collect data for prescribed drugs, and provide insurance
information [12,13]. For handling COVID-19 data, the IoMT collects important
data of patients, provides disease symptoms and trends, automates process logistics,
facilitates remote caring, and gives patients access to control their healthcare [14–
16]. Hence with one or more of these devices, it is possible to monitor patients in
real time, enable treatments, conduct check-ups, and even follow up on the progress
of improvement. In turn, this helps to deliver faster healthcare while reducing
hospital costs. With sensors in medical devices, data can be collected and securely
sent over a wireless network to a health specialist who would then analyze the data
for possible anomalies. Essentially, this has a higher potential to manage the
COVID-19 pandemic and reduce the spread of the virus more securely and reliably
than conventional strategies employed to address the widespread epidemic
experienced in the past. This is evident in the operation that expands technology
capabilities while minimizing the risk of exposing patients, medical practitioners,
and health specialists to COVID-19 since contact is only established on a critical
need basis. Patients can also improve their self-care and easily access support using
digital and communication tools developed using blockchain technology [17]. An
illustration of the operation of blockchain technology in the management of
healthcare and sensitive data using through the IoMT is shown in Figure 6.2.
In content and focus, this chapter explores how blockchain technology can
transmit medical information that comprises sensitive data associated with the
COVID-19 pandemic more securely and efficiently. This would highlight the
essential features of blockchain technology, potential challenges, and emerging
opportunities related to healthcare data of the prevailing COVID-19 pandemic.
Accordingly, the key contributions include recent developments of blockchain
technology in relation to its application to handling medical data, potential benefits,
key current and impending challenges, prospects of blockchain technology,
research efforts on blockchain technology in healthcare, blockchain technology
application to handling healthcare data, and real-time COVID-19 pandemic data
analysis. The conclusion of this chapter reiterates key findings, future research
directions, and further reading.
6.2 Recent developments in blockchain technology
In recent times, more attention has been given to blockchain. As it advances in

several applications such as healthcare, banking, manufacturing, computer net-
works, communication, and governance, there has been an increasing acceptance
globally [18]. The popularity of blockchain technology rose from its capability to
improve service security, efficiency, and scalability in applications. This has made
Patient medical
Treatment
history
checklist
Blockchain
Distributed
Cloud
ledger
storage
Lab
results
Figure 6.2 An illustration of the management of healthcare data using blockchain

technology
it a top choice for incorporating decentralized ideas and procedures in various

domains. Since 2019, blockchain technology has developed into a strategic industry
that can be integrated into cross-chain, open sources, and various industrial appli-
cations [19]. In these applications, blockchain provides distributed data handling of
protected and time-marked records of information, which could be sensitive like
patients’ personal information in healthcare. A typical example of this is shown in
Figure 6.3, where the data of users are stored on a distributed ledger in each
blockchain, which is all directly connected to a common network, without an
intermediary.
In this case, blockchain combines separated health records on the network to
track the personal health progress of an individual. Although the first blockchain
technology named Bitcoin was introduced in 2009 [19], several unexpected events
have occurred over the years that have given blockchain technology an upward
trend. This intensified with the emergence of the COVID-19 pandemic, with high
expectations of the blockchain domain beyond 2021, as statistics reveal an
Distributed ledger
Blockchain
User 2 Distributed ledger

Blockchain
User 1
Distributed ledger Blockchain
Blockchain network
User 3
Distributed ledger
Blockchain
Distributed ledger
Blockchain User 5
User 4
Figure 6.3 An example of blockchain technology and data stored on the

distributed ledger
Global market size of blockchain technology (million USD)

2,500
2,000
Value (million USD)
1,500
1,000
500
0
2016 2017 2018 2019 2020 2021
Year
Figure 6.4 Global market size of blockchain technology (2016–2021)
estimated 1 000% increase in the size of blockchain technology from the year 2016
to the year 2021 worldwide [20], as shown in Figure 6.4.
The application of blockchain technology in the handling of medical data is
constantly transforming healthcare. As new measures are developed to manage the
spread of COVID-19, this has been an invaluable tool to medical experts, health-
care specialists, policy makers, and patients. In these times, digital transformation
is more than just a choice for any business. It has become essential for survival in a
rapidly changing environment with limited resources, as the COVID-19 pandemic
increasingly puts a strain on everyday companies. In healthcare, blockchain tech-
nology is making remarkable changes in managing health issues and the provision
of care. In effect, the sector is looking to become more digital with blockchain help
in data exchange, supply-chain management, and contracts [21].
The autoregression integrated moving average (ARIMA) model with machine
learning is another recent development that can be integrated into blockchain
technology. Using the available sensitive data, this model contributes to COVID-19
data handling by forecasting its outbreak or predicting the reported cases of
COVID-19 in a certain region [22]. It does this through phases of transformation,
model identification, estimation of the best suited ARIMA, and diagnostics.
6.2.1 Healthcare data systems

The control and management of the COVID-19 pandemic have seen a consistent
development of healthcare data systems. They combine the generation, access, and
storage of an individual’s medical information digitally. They are designed to
operate in real time as a patient-centered record that authorized users can access
instantly [23,24]. This has helped to manage the spread of the pandemic better since
it includes a broader view of the healthcare condition and exposure of each
COVID-19 patient. These systems are developed by blockchain technology that
further helps authorized clinical specialists remotely create, manage, and share
medical information across multiple health institutions. Another form of these
systems developed recently is an optimized version of Personal Health Records,
where patients can control who has access to their medical information and
remotely manage and track their healthcare, regardless of the healthcare institution
providing their care. The recently developed healthcare data systems combine both
functions, which can address a health situation at a given time, where more sensi-
tive sensors are used on healthcare devices or in health institutions, and patient-
centered healthcare is provided [25,26]. For COVID-19, this may involve infected
and noninfected individuals.
An individual’s healthcare data can be generated from medical devices at a
different location or while that individual receives the provision of healthcare ser-
vices in a medical institution. However, the dire need to implement optimized
blockchain technology in the handling of medical information systems during the
COVID-19 pandemic arose because most of the systems before the advent of the
pandemic were not interconnected, leaving critical players in healthcare provision
with restricted, fragmented, or compromised information, despite having their
private database [27]. In response, new health blockchain architectures have been
developed, with others underway. A general scenario of these architectures in
relation to securely handling data is shown in Figure 6.5.
In this architecture, the blockchain stores medical information in a digital
electronic format. The health elements stored have network applications and
Health entity
(internal system)
Database
server
Synchronization Internal work

stations
External authorized
users
Application
server
Blockchain
Health entity
Web server (internal system)
Database
server
Synchronization
Internal work
stations
Application
server
Figure 6.5 Blockchain architecture in healthcare
database servers connected to the blockchain to store and process this medical
information. Through another application server within, authorized clinical per-
sonnel can communicate with the data system. At the same time, the medical
information from the databases of medical devices in their institutions is combined
with that of the blockchain, using a synchronizing component. This setup guaran-
tees a high level of security and data integrity in the event of breakage or con-
nectivity loss in the chain [27].
Systems for ensuring that previous analyses are not used for new diagnostic
processes have also been developed to support data handling. While these systems
have been developed over the last decade, it has also been accompanied by chal-
lenges such as incapacity for early detection of disease, poor data quality, high cost,
patient privacy, lack of interoperability, and inefficient patient data management
[28,29]. The blockchain-developed eHealth system of today improves the system’s
intervention by supporting electronic medical information solutions that overcome
the limitations of power while synchronizing clinical research and information
technology. In effect, more privacy, better scalability of data, and less fragmenta-
tion are experienced. Blockchain technology maintains integrity by securely storing
data that otherwise may be inappropriately accessed due to the instability of com-
munication networks or intermittent connection failures [27]. Other healthcare data
systems developed using blockchain technology are being applied to medical
records of several individuals at the risk of exposure to COVID-19 for swift
intervention and addressing connectivity and data transmission issues.
6.2.2 Healthcare data exchanges

As the number of patients infected with COVID-19 increases, the provision of
adequate medical care becomes a challenge in the area of diagnosis, monitoring,
treatments, and cost [30]. This is particularly because healthcare services cannot
securely send and receive sensitive data across trustworthy channels to improve
their services in the overwhelming times imposed by the pandemic. However,
recent reports show that healthcare providers have adopted blockchain technology
to enable network access among medical practitioners, patients, and healthcare
specialists while maintaining the security of data and the integrity of the system.
This comes in designed products that give access rights and manages permissions
of medical information exchanges. In contrast, others manage the information of
the medical provider in terms of storage and transmission [31]. In order to reduce
the load of reported cases, a few health groups are also coming together to ensure
an up-to-date directory where essential information is shared among themselves
and accessible from a database. Additionally, blockchain technology is used to
develop a credential verification system that allows medical practitioners at the
forefront of COVID-19 pandemic cases to prove their license for the operation [32].
Although in the past few years, the quality of healthcare has improved with
wearable devices, implantable devices, and the IoMT [33], blockchain technology
helps to authenticate and secure healthcare data that may be otherwise accessed
inappropriately [30]. This is highly instrumental, as securing patients’ personal
information requires healthcare systems to meet the unique requirements of data
transfer and interoperability, which the concept of decentralization achieves
excellently.
6.2.3 Healthcare administration

Following data-handling standards, including privacy, security, transmission,
exchange, content, and terminology, blockchain technology is employed to speed
up some healthcare processes and reduce transaction costs [34]. The rise in the
spread of COVID-19 accompanied an urgent need to exchange data through
healthcare systems efficiently. To combat the pandemic, healthcare providers,
health policy makers, health insurance companies, and healthcare institutions col-
lectively understand the complete picture of the pandemic through the exchange of
essential data. These are only achieved by implementing standards that guide the
capture and transfer of medical information, where blockchain technology sig-
nificantly impacts today. Typically, patients’ medical information is grouped to
form a dataset to evaluate the quality of care provided, measure healthcare out-
comes, and report on the safety or progress of health [35]. Prior to the active
integration of blockchain technology in healthcare and the inception of the COVID-
19 pandemic, healthcare support was more complex, costly, and slow, which
invariably generated a lower quality of healthcare provided and an incapacity to
manage the ranging spread on COVID-19 when it was first discovered.
Although these healthcare standards and their administration in the provision
of healthcare have been in place before the pandemic, the need and challenges have
only intensified with the COVID-19 pandemic. Following the required procedure
and consent, several representatives work to enforce the standardization of sensi-
tive data transactions, which, for the most part, requires access to complex data
through a high level of transparency, confidentiality, and integrity. Intelligent
Healthcare Networks are being developed to track data gathered, transmitted,
received, revenue, and permission status using blockchain technology. As expec-
ted, this comes with an improved scalability and speed of health administrative
process that can also be employed in a larger situation like the fight against the
spread of COVID-19.
6.2.4 Pharmaceuticals
Pharmaceuticals and drug production play a significant role in managing the
COVID-19 pandemic, and the industry is fast moving toward blockchain technol-
ogy. Given that large data are a big part of pharmaceuticals, the management of
data is essential. Aspects of the industry that may generate a large amount of data
include pharmaceutical research for the production of vaccines and suppressants;
clinical testing; development of applications for healthcare like the IoT; and the
expansion of new regulations for drugs and supply chain. As large as these data
may be in many cases, it must also be maintained and organized. Blockchain
technology is currently improving this and providing smart security of data and
transparency in data handling and logistic activities [36]. In effect, safer drugs
production can now be enforced with minimum errors that can be detected and
traced, if any. This is important to deliver quality healthcare and minimize unsafe
drugs, especially in curbing the spread of COVID-19. Accordingly, several initiates
are currently being developed to support this. Some of these projects track and
verify pharmaceutical returns using blockchain technology that detects when a drug
is touched and the person who did. This is expanding to other applications as in the
year 2020, blockchain technology helped to combat the production of counterfeit
COVID-19 drugs, while sophisticated tools like machine learning are employed to
optimize this function in this present time [37,38]. This would require training deep
learning models using the interpreted COVID-19 dataset within the network for
data-transfer learning and data augmentation [39].
With the increase in the number of individuals vaccinated against COVID-19,
blockchain technology is implemented to accurately manage medical information
on the storage and transportation of several batches of vaccines. This is done using
the IoT through sensors affixed to the storage containers and transport vessels of
the vaccines. While helpful data are collected, transmitted, and stored on the
blockchain, healthcare institutions, regulators, and distributors can instantly view
this information to affirm the safety and effectiveness of how the vaccines were
stored and transported. This has dramatically minimized the spread of COVID-19
and maximized clinical research and trials. These require the careful handling of a
large amount of sensitive data from various sources. Blockchain functions to con-
nect these disparate data from different clinical research facilities and then create a
traceable record with an in-built safety mechanism and an audit framework
Manufacturer Wholesaler Health Pharmacy Patient

distributors
Figure 6.6 Blockchain technology healthcare supply-chain management system
accessible to regulatory bodies and healthcare services. The basic structure of how
blockchain technology securely handles data through a supply-chain system is
shown in Figure 6.6.
In order to carry out this function for data security, blockchain technology is
developed to instantly alert authorized individuals when any attempt to alter data at
any point of the chain is detected. In compliance with regulatory requirements, this
has improved the authenticity, confidentiality, and trust in the data from healthcare
systems currently managing the COVID-19 pandemic.
6.3 Potential benefits of blockchain technology in data

handling
Blockchain technology provides healthcare data systems with adequate access

control, data security, data integrity, auditing access, and data sharing [40,41]. It
includes the security of sensitive data in a supply chain by leaving all added data to
the blockchain unchanged once they are added. This has the potential to automate
data exchange log instances and policies on data access for subjects and profes-
sionals in healthcare to access and verify at any time. Given that the COVID-19
pandemic has strained healthcare systems globally, and healthcare professionals are
pressured to control the spread of the pandemic, relevant data access must be
adaptable to changing healthcare conditions. In handling these data securely,
blockchain technology may help integrate COVID-19 pandemic data generated
from different health service providers and several healthcare intermediaries [42].
Such systems may also store patients’ health histories, which can be useful to
health service providers, pharmaceutical companies, and medical product con-
sumers. In this case, blockchain technology may offer the advantage of establishing
interoperability. The potential benefits of blockchain technology in healthcare can
be summarized in Figure 6.7.
Healthcare systems use different approaches to prepare for unforeseen occur-
rences like the COVID-19 pandemic and medical emergencies in mitigation,
recovery, and response. In critical situations, blockchain technology tools provide
the benefits of cooperation and sound communication between critical elements of
Electronic
Implants medical
Ambulatory
Health records
medical
solution Smart health
devices
devices
Activity
monitoring
Connected
Peripherals equipment Lab on a chip
Home medical
devices
Health Hand-held Hospital

software Wearables medical tech medical
applications devices devices
Patient generated data Clinical data
Figure 6.7 Potential benefits of blockchain technology in healthcare
healthcare provision like finance, logistics, health workers, back-end data input, and
management for decision-making [43]. Generally, this improves data integrity,
reduces medical costs, and decentralizes operations. Blockchain also can deliver
precision that is essential to improve the care of patients and management of critical
health situations. At the same time, records of medical information are connected
across networks. Using the IoMT to track the different data nodes and update patient
data, these benefits consolidate real-time monitoring of health and the measures
adopted to inhibit the pandemic. Globally, blockchain technology adds benefits of
high data accuracy and reliable data security when health systems are developed
based on seamless data sharing [44,45]. Hence, blockchain technology can be used in
any data-driven areas of healthcare, including health claims, supply chain, healthcare
records, and supply chain. Top-rated potential benefits of blockchain technology in
securely handling COVID-19 data may be seen in the following.
6.3.1 Better exchange of healthcare data records

In the exchange of health records, blockchain technology can potentially address
interoperability challenges by securely distributing healthcare data [31]. A big
challenge of achieving healthcare data interoperability is the inadequate capacity
for healthcare data systems to share their healthcare data, especially during a pan-
demic like COVID-19. While this may vary from one health facility to another,
clinical specialists or specific doctors at the forefront of combating COVID-19
need sensitive data that must be recent in most cases. However, health services
often experience redundancy that delays the care delivery process. Nevertheless,
blockchain technology helps to easily manage this easily by creating a seamless and
secure access to healthcare data and treatment records [46].
6.3.2 Validating trust in medical research and supplies

Health facilities that have handled COVID-19 cases experience challenges in
shifting health results and accessing reliable data. This makes it difficult for them to
maintain and expand their capacity to provide adequate care for patients and con-
trol a critical situation like the COVID-19 pandemic. Blockchain technology has
the potential to enable the secure transfer of valuable records of research outcomes
and clinical trials, which minimizes the possibility of errors or interference in
healthcare data records. This is possible because of the encryption associated with
blockchain. In effect, previous clinical trial logs can be tracked by blockchain while
avoiding redundant clinical trial outcomes or data that have been altered [31].
Given that this area of healthcare management is guided by manufacturing, dis-
tribution, and administration regulations, the government may provide the needed
assistance for health facilities to meet the challenges of the COVID-19 pandemic.
The fight against COVID-19 has led to several clinical trials to develop a
partial remedy for the virus. This is known to yield so much data that those may be
challenging to track. These data may also be hidden, corrupted, forged, or erased by
external influences and some researchers. In the consideration of this in handling
COVID-19 pandemic data, blockchain technology has the potential to confirm any
document transmitted through the network authentically. Blockchain may first
maintain an inalterable record of data that can all be verified for approval [47]. In
turn, the transparency and accuracy of these data build confidence in researchers,
policy makers, and medical supply firms.
6.3.3 Validating correct billing management

Blockchain technology offers a high-integrity option for the tracking and refreshing
of data at a given time. The correctness of billing can be validated by improving the
auditability and security of billing while supporting seamless billing processes and
efficiency and avoiding theft. An integration of blockchain technology to handle
this kind of healthcare data has the potential of adequately detecting fraud and
related activities [48]. It simplifies billing tasks and engages with sound network
access to adapt to healthcare demands that are constantly changing, particularly in
the COVID-19 pandemic. Apart from fraudulent activities, the complex yet a large
amount of data from the pandemic and healthcare field, in general, may also gen-
erate financial mishaps. However, recent blockchain technology developed can
improve billing using automated paying processing based on the data received.
6.3.4 Internet of Things (IoT) in healthcare

The use of blockchain technology in handling COVID-19 data provides the
potential benefits of tracking emergency room occupancy, inventory, patients,
availability of medical devices, and health progress. This is improved with the IoT
that are the key elements that capture and transmit data to an authorized clinical
destination. The pandemic has driven more visits to health institutions more than
ever before, with more in-hospital stays and beds occupied. Hence, responding to
health issues and tracking patients may be difficult for doctors around the world.
IoT in blockchain technology can make these responsibilities easier by providing
access to secure and relevant data when needed [49]. These benefits can also be
extended to temperature control in the rooms occupied by patients and medical
supply availability. Predictably, several technology firms are developing IoT-based

blockchain technology systems that would support adequate healthcare in the pre-
sent uncertain times of the COVID-19 pandemic.
As an advantage, these systems work in a distributed and decentralized way,
maintaining data that cannot be altered and, therefore, not misleading. With a good
setup of the IoMT, fraudulent activities could also be detected easily because
blockchain allows one to efficiently track the source and prescription of drugs at
any time within the network. Optimizing blockchain IoT technology with machine
learning could enhance these benefits and make medical operations of controlling
COVID-19 more efficient.
6.3.5 Optimized privacy and data security

For the effective management of COVID-19, collaboration and coordination must
be established between health facilities, health specialists, and authorities in dif-
ferent regions [50]. This would be impossible with data security and integrity
issues, which has been the case before now. Although sufficient data are constantly
exchanged in managing a global pandemic, those are often not trusted due to the
data system inconsistencies, giving rise to experiences of delayed interventions,
inaccurate diagnosis, and healthcare errors. However, through private and public
chains, blockchain technology presents the potential benefit of an authorized data
access control and the control of cyberattack threats that limit data distribution
while maintaining interoperability standards. For a strong security of data and
improved privacy, the private blockchain encrypts sensitive medical information
that is only accessible by authorized users.
Data privacy and security are also demanded in the flow of healthcare services
overwhelmed with insurance, medical claims, and finances issues, which can
expand to lose and fraud running into outrageous sums of money. This can be
experienced because of the poor transparency of data that are incomplete or
inadequately stored. Blockchain technology may recover lost funds, monitor care
received by patients within and outside a health facility. With blockchain tech-
nology, medical events and clinical operations can be evaluated from the data of
care that are provided from the blockchain across the network. Accordingly, sys-
tems that enhance data transparency are being developed to control the spread of
the COVID-19 virus better.
6.4 Key challenges of blockchain technology in data

handling
Implementing blockchain technology in healthcare applications like secure hand-
ling of the COVID-19 pandemic data comes with some limitations that prevent
meeting the goals of managing the spread of COVID-19 in part or totally. Key
challenges identified include security, speed, interoperability, stringent data pro-
tection regulations, scalability, and privacy [51].
6.4.1 Security
While blockchain technology provides data encryption, there are possibilities of
revealing the identity of individuals or other sensitive data, mainly when separate
medical information of an individual is linked within a network, and a public
blockchain is adopted for healthcare data [52]. Some health facilities have also
experienced attacks on their healthcare blockchain from fraudulent organizations,
and this breaches the security of sensitive data leading to a compromise of privacy.
This may also be due to a compromised data decryption and encryption function in
the blockchain, which these organizations leverage by gaining unauthorized access
to sensitive healthcare data stored. In most cases, this may not be detected early on
if the blockchain is not maintained for optimized functions. This is a significant
challenge as several reports of cyberattacks on blockchain networks control the
data functions of cryptocurrencies [53].
6.4.2 Speed
In many cases, the performance of the blockchain technology system is compromised
with a low processing speed. Although the blockchain may have great potential
benefits for combating COVID-19, inadequate data-processing speed is considered a
major challenge [54]. For health cases like the COVID-19 pandemic, more data
should be processed at high streaming speeds in data collection and transmission to
the appropriate destinations. This is because processing in blockchain technology may
introduce some latency, particularly in cases where all network nodes participate in a
data validation process during and after blockchain setup [53]. In addition to the data
load, this results in a significant processing delay. The tradeoff in improving the speed
may consider a few nodes in the data validation or consensus processes, which may
not be sufficient to deliver an optimized data function of the size of data generated
from the COVID-19 pandemic. While the public blockchain may require any node to
participate in these data processes, the security and content of data transmitted may be
inadequate to meet the need of managing the pandemic.
6.4.3 Interoperability
In order to manage the COVID-19 pandemic, it can be challenging to implement
blockchain technology where there is no existing standard for developing secure
handling of data. This makes interoperability a major challenge for developed appli-
cations that securely handle a large amount of data. There is a lack of interoperability
between two blockchain systems, so the exchange of relevant data becomes difficult
[54]. In most cases, it is impossible for one platform to synchronize with another. This
results in loss of data, inaccurate data, or delay in obtaining data. In turn, this gen-
erates an inconsistent provision of healthcare and control of the pandemic, which is
both critical end-products of handling the pandemic data securely.
6.4.4 Stringent data protection regulation

The General Data Protection Regulation (GDPR) supports the right of users to
request a total erase of their data. This is against the operations of blockchain that
leaves data unaltered once it has been saved. As a result, implementing blockchain
technology to handle data securely can be challenging given the sensitivity of the
COVID-19 data pandemic. Although, in some cases, this regulation may not be
considered when other people have to make data privacy decisions for patients who
are unable to do so, historic data for medical record purposes may be altered when
data protection regulations are effected. Nevertheless, this would be counter-
productive when erasing a medical history is desired by patients or their repre-
sentatives. In turn, it would be difficult to control the spread of COVID-19 and
develop future preventive measures with accurate data.
6.4.5 Scalability
This is another major challenge when implementing blockchain technology in
secure data handling. The scalability of blockchain technology data systems is the
capacity to handle the changing demands of handling data resources. While scal-
ability requires specialized techniques to deliver as it should, it is a significant
challenge in healthcare solutions because of the high volume of data handled in the
COVID-19 pandemic. Generally, the blockchain was built with the capacity to
store a too high volume of medical data than it can take, leading to severe perfor-
mance degradation [6]. Given the conditions of the COVID-19 pandemic, the
traffic on the blockchain would become bulky with an increasing number of
transactions daily. As each node stores all validated data transactions, a restriction
would be experienced on the size of the block and the time taken to create a block
[54]. In turn, there would be a delay in validation and data processing since the
management of the pandemic is accompanied by a high data load [55]. Hence, the
challenge of scalability also affects the speed of transactions on the network.
6.4.6 Privacy
As key data balances and details are validated by authorized network users,
blockchain technology may have privacy leakages. Although blockchain technol-
ogy has the potentials to offer patients and clinical specialists better control over
the healthcare data of individuals, which translates to better management of the
COVID-19 pandemic, it is not originally designed for this function, so at an initial
stage, sensitive data can be stored on a distributed log that is accessible to any user
on the network. This limits the privacy function of blockchain technology, espe-
cially with a weak network. Hence, the privacy function of blockchain technology
can be supported when the blockchain network is developed to have nodes that
process data without revealing any of it. This may be a good measure to handle this
challenge, but more resources would be demanded optimum function.
Nevertheless, systems that provide this optimization service to blockchain tech-
nology have been developed with newer systems underway. However, they offer no
guarantee of perfection with the optimized networks.
While implementing blockchain technology in healthcare applications persist,
some solutions have been developed to manage them. In order to prevent fraud or
unauthorized access to data in the blockchain, a mixing solution system is
proposed. It transfers a given fund to an output address from multiple input

addresses to maintain data anonymity in the network. To improve scalability,
encrypted healthcare data can be stored outside the blockchain, where condensed
medical information from the dataset and an access guide are both stored on a
blockchain at a given time [56,57]. This also provides a solution to the stringent
data protection regulations as the stored data can be deleted while leaving data
pointers on the blockchain. The tradeoff in this is the partially lost redundancy that
may enhance data availability in the blockchain. From a legal perspective on data
handling, the concerns that arise in the COVID-19 pandemic are storing, sharing,
and accessing data from the distributed database of a blockchain network [54]. This
is owing to the data laws and policies that invariably limit the function of block-
chain technology. The solution to this would be to revise these policies by inter-
national health organizations, policy makers, and country leaders, which may have
to introduce new health policy regulations on digital health, digital connectivity,
data sharing, and the digital divide.
The security of healthcare data and privacy protection issues can be resolved
using a permission consortium blockchain for digital healthcare applications like
the secure handling of COVID-19 data. Most of the threats to security may also be
contained by applying known security measures to develop software and codes to
optimize blockchain technology functions. This also helps to reverse traces of
fraudulent transactions. Smart contracts of the blockchain technology can also help
to define and program rules that may help to control the behavior of the blockchain
in the data handling of COVID-19 patients.
6.5 Prospects of blockchain technology

Blockchain technology can assist in modifying the current healthcare data–hand-
ling systems to combat the COVID-19 pandemic in more productive ways. The
current systems have adopted efficient measures to manage the global condition,
which is evident in the rollout of vaccines that would help to manage the effects of
the virus in infected individuals [58]. However, these potential improvements of
blockchain technology are expected to yield better results that would be evident in
a progressive decrease to the number of COVID-19 cases, which would also inhibit
the spread of the virus to a minimum than ever before. The major areas of
improvement are summarized in Table 6.1.
These prospective areas of improvements by blockchain technology are
already being explored, and several ideas have been developed to support the
efficient control of the COVID-19 pandemic. Noteworthy is the role of elements in
a blockchain-based medical information management system, which are important
to convey certain functions in implementing blockchain technology. These block-
chain elements include storage memory, cloud storage, distribution ledger, and
hardware. All these have their functions in the overall impact of blockchain tech-
nology in healthcare. A displacement of any element in the data-handling frame can
potentially compromise the blockchain technology operation. A summary of how
Table 6.1 Prospects of blockchain technology in data handling
S. Blockchain technology Improvements to the healthcare data handling system

no. prospects
1. Protection of digital Digital healthcare systems are susceptive to cyberattacks
healthcare data systems globally, leading to a compromise of sensitive medical
information [59]. This mostly occurs in a compro-
mised network connection. When concerns on security
and privacy are addressed, digital healthcare data
systems will improve. Blockchain technology can help
provide the highest standards of privacy protection,
trust, security, and optimization in these systems,
leading to better protection and privacy of data.
Further to this, an exchange of data can be optimally
established without an intermediary, enabling the
storage of all patients’ healthcare history and treatment
records securely accessible, decentralized, and trace-
able. However, where there are limited resources, the
cost of implementation may be increased
2. Healthcare invoicing Fraudulent activities are common in the healthcare
billing system [60]. More so, the financial transaction
processes are time-consuming and require more
resources. While measures like complex codes pro-
gramming help structure healthcare billing systems,
they may also come with delayed inaccurate, dupli-
cate, or unintentional billings [10]. Integrating pro-
gram codes in blockchain technology systems can
optimize healthcare billing. It could help make the
billing and payment process more secure and easier
than the conventional billing systems in most health-
care facilities. For example, a delay in paying the bills
of insurance claims is often experienced. An imple-
mentation of an automated blockchain technology
system can speed up invoicing and payment processes
of insurance claims by storing data in a manner that
cannot be altered. This decreases the excess resources,
saves time, and helps insurance providers make
payments faster
3. Drug traceability There have been cases of detected counterfeit drugs
entering the market in the supply-chain cycle for the
distribution of drugs for healthcare purposes [61].
From the manufacturing to the distribution phase, the
supply of fake drugs has become a business for many
years and is estimated at several billions of dollars
globally [62]. As a result of this, it can be deduced that
the pharmaceutical industry is weakly protected.
Blockchain technology can help track the manufac-
turing of these drug chains since blockchain stored
data transactions cannot be altered. According to
business needs, blockchain technology offers phar-
maceutical companies the option of employing private
(Continues)

no. prospects
or public blockchain systems that can help them obtain
a complete clinical trial of a drug. When the drug is
touched or moved, this information is stored as
unalterable data on the blockchain database. These can
significantly improve drug traceability and minimize
the risk of supplying and distributing fake drugs
4. Healthcare data manage- COVID-19 healthcare and the management of the
ment pandemic require access to a patient’s full medical
records to provide patient-centered care that would be
different from another patient. Handling these data can
be challenging because healthcare data systems are not
equipped with the needed privacy, security, and trust
that the sensitivity of the data demands [63]. In most
cases, this data may also be altered or deleted, thereby
incurring an extra cost when patients need to move
from one health facility to another since there may be
no records of tests and medications from past
healthcare provided. Blockchain technology has the
potential to overcome this challenge, owing to its
decentralized peer-to-peer network structure, which
can only be accessed using smart platforms [64]. In
addition, the record of healthcare provided is stored on
several blockchain nodes within the network. This
way, data transfer can be done from one health facility
to another securely and transparently. Blockchain
technology can also help to reduce cost since patients
would not have to repeat some clinical tests or
diagnostics that have been done earlier
5. Precision medicine Precision in medicine and the outcome of clinical trials is
an essential part of healthcare monitoring and provi-
sion. The process of clinical research and trials can be
optimized with blockchain technology by potentially
addressing incorrect clinical trial data and supporting
data integrity [65]. Often, inaccurate clinical trial
outcomes lead to inefficient interventions, and for a
case like the COVID-19 pandemic, this should be
encountered. With a programmed blockchain technol-
ogy system, more transparency can be provided to
improve analytic data accuracy. Other areas of
precision medicine that can be potentially improved
with blockchain technology are clinical supply audit,
reducing clinical trial time, recruiting patients, and
tracking prescriptions. It may also be used to control
genomic sequences [66], which may be used to treat
different genetic disorder-induced illnesses and infec-
tions. Blockchain technology can capture the genetic
data of an individual that would help generate more
(Continues)

no. prospects
accurate results. More so, the interoperability function
in most blockchain technology systems ensures that
individuals can control their own data while it is being
shared between health organizations. With this, there
would be no need for a central database managed by an
intermediary but susceptible to cyberattack
6. Data access permission Permission to access the medical data of a patient may be
required in emergencies. When this is inconsistent,
human life may be in danger due to the limited access
to relevant medical information that can assist medical
professionals in providing the needed healthcare. For
secure and seamless data access permission, block-
chain technology has the potential to create a contract
consisting of set rules that grant access to data through
smart platforms. These contracts could be designed to
include an automated schedule of different workflows
and conditions for specific cases, like when two
medical specialists need access to private data at the
same time for the provision of specialized healthcare.
Access control can be managed by the patients
themselves using cryptographic master keys, which are
entered to unlock healthcare data within a preferred
length of time [10]. With this, a patient can share any
data with healthcare facilities or clinical specialists as
needed. In emergency cases where a patient is
indisposed to grant access, blockchain technology
systems can be programmed as robots for certain
functions, which may assist in saving the life of a
patient. Thus, access privileges of reading and writing
can be added to the blockchain. All these prospects can
help minimize human-imposed errors and save time
needed to collect patient data
blockchain is used to access and transfer data stored and processed by blockchain
elements in a typical healthcare scenario is shown in Figure 6.8.
6.6 Research on blockchain technology in COVID-

19 healthcare
The ongoing research on blockchain technology in COVID-19 healthcare has
revealed innovations that support a better quality of healthcare and inhibit the spread
of the pandemic. Identified areas of these researches include pharmaceuticals,
healthcare data processing, billing, cyber security, and healthcare data system opti-
mization. Some of these studies have already tested and implemented outcomes in
Support intervention Valid

decisions prescription
Doctor Pharmacist
Health Data collection

advice and analysis
Patient Blockchain
Data analyst
Validated policies Help in disease

and regulations diagnosis
Healthcare X-ray
policy maker specialist
Figure 6.8 Blockchain technology-enabled healthcare data management
real-time health applications, while others may be undergoing revision. In relation to

the prevalent COVID-19 pandemic, health research outputs help provide evidence-
based healthcare, treatment plan, and health prevention options [67]. They also guide
doctors and other clinical specialists on the best treatment options for their patients.
New healthcare tools, the development of medicine, the adoption of new care pro-
cedures, effective interventions, and preventive disease strategies are all possible due
to constant clinical research outputs [68]. In turn, healthcare becomes more efficient,
personalized, and convenient. A few recent research works on blockchain technology
in healthcare applications is highlighted as follows:
The blockchain-network Electronic Medical Record system is created as a pro-
posed solution for the security, privacy, and fragmentation of medical information
from different databases, limiting effective treatment and disease prevention that
arise from issues like data leaks, data loss, or misinformation [27]. This system was
developed to prevent the direct risk that these issues pose on individuals and to
reduce public health costs. The proposed system may be vulnerable to connectivity
failures, leading to system data inconsistency, especially in developing countries. In

response, a blockchain technology architecture called “HealthyBlock” is presented as
a unified Electronic Medical Record tool that has resilient data integrity when con-
nection fails and characteristics such as privacy, security, and usability. The eva-
luation of this system shows high efficiency in updating, unifying, and securing
medical information of patients across any clinical network provided connected.
Blockchain technology in healthcare is also validated with suggestions of
clinical practices that uphold COVID-19 safety. Accordingly, the combination of
blockchain and artificial intelligence systems is proposed to create predictive tools
that are able to contain the COVID-19 global pandemic risks [43]. This response to
the adverse effect the COVID-19 pandemic has on risk management. This system is
designed to be a strategic tool that strengthens operative protocols while creating
the basis for effective medical decision processes. The limitations in its adoption
for managing the COVID-19 pandemic were highlighted in a SWOT analysis,
which also presented top future digital healthcare opportunities. Two of these
limitations were high operation cost and scalability for more data storage on
available servers. Given that measures to manage both challenges have been pre-
sented in several research outputs, the proposed system is evaluated to improve the
clinical practices that are COVID-19 safe. This was supported in relation to clinical
workflow concepts from various blockchain technology models.
The prospects and challenges of adopting blockchain technology in tele-
medicine and telehealth are explored in relation to the role of blockchain in the
provision of privacy, security, operational transparency, fraud detecting trace-
ability, and unaltered health records, to inhibit the spread of the COVID-19 virus
[9]. This solution is proposed to manage the scarce resources in healthcare so as to
control the burden of managing COVID-19 patients in health facilities. Although
there are concerns about issues related to patients’ insurance claims and the ver-
ification of credentials, the evaluations of the proposed solution reveal that inte-
grating blockchain technology improves telehealth care services by providing
remote, decentralized, reliable, and trustful health services. The further assessment
shows good security and accuracy in detecting faults in medical test kits for diag-
nosis outside health facilities. As a future development area, the global adoption of
blockchain technology in healthcare was encouraged if the challenges identified in
the implementation process are resolved.
Another solution is modeled for sharing, analyzing, and organizing COVID-19
data with interoperability and privacy-preserving methods. This is aimed at
improving the time, cost-effectiveness, and outreach for treatment and disease-
control interventions. Accordingly, a blockchain system is proposed as a promising
technology to drive effective disease control and monitoring [4]. For a disease like
the COVID-19 virus, a health surveillance system is required for regular diagnosis,
control, and treatment interventions. This solution was presented because of the
blockchain features such as data immutability, decentralization, cryptography, and
data provenance. In effect, a literature bibliometric analysis of integrating block-
chain technology in the data management system of a disease like the COVID-19
virus was performed. The suitability of many existing blockchain platforms was
also evaluated in relation to disease data storage, processing, and sharing. Also
addressed were optimization models on the blockchain application, like the design
of machine learning models that preserve data privacy better, which can also sup-
port collaborative learning. Overall, a disease data-handling system that integrates
machine learning models on blockchain was recommended [4].
Other studies identified new characteristics of blockchain that are of great
benefit to healthcare systems, including healthcare industry collaboration, health
business models, optimized auditability, disintermediation, and continuity [18].
Results from statistical models identify blockchain technology as a platform also
created to provide reliable recording, where initially fragmented medical informa-
tion records would be combined to easily track the health records of an individual
[69,70]. The difficulty of building a tracking capable foundation of high integrity
has led to an inconsistency in accessing healthcare records. Further to this, history-
based diagnoses are costly because of the complexity of medical information
records, privacy issues of the network, and fragmentation. Blockchain technology
offers a system to address all these through the continuous track of services, which
it provides [18]. The combination of blockchain and artificial intelligence can also
offer a solution for these healthcare data problems while also supporting the
seamless access and storage of data on the blockage. However, these ideas can be
implemented when regulatory barriers from data privacy policies are addressed or
revised to deliver healthcare or mitigate COVID-19 spread.
While many more research outputs have been presented, it is worthy of con-
sidering collaboration in healthcare to drive the development of brilliant ideas and
effect the needed change. In managing the COVID-19 pandemic by exploring
secure data handling, this is highly critical considering that a large amount of
medical data are shared among healthcare facilities and clinical experts expedi-
tiously across several miles globally.
6.7 Real-time analysis of COVID-19 pandemic data

The global situation of COVID-19 pandemic cases is explored in statistical terms to
give a broader perspective of integrating blockchain technology to the secure
handling of data. Across the globe, there have been a rise and fall in the number of
infected individuals recorded. Therefore, a regular analysis of COVID-19 patient
data is critical for understanding clinical progress, measuring how effective the
strategies for mitigating the spread are, potential effects of treatments, and expected
outcomes [71]. Accordingly, adequate statistical models are necessary to affirm
data in health facilities at a given time. In-line with this, real-time clinical examples
are used to evaluate standard statistical models and their mathematical framework
regarding real-time COVID-19 pandemic data.
6.7.1 The susceptible recovered infectious (SIR) model

This is a simple model used for the analysis of COVID-19 pandemic data, where
the population is divided into three parts: the susceptible but not infected, the
infected, and the recovered (or deceased) [72]. Several research outputs have been
generated by modeling the outbreak using the basic susceptible recovery infectious
(SIR) model. Examples include the estimation and simulations of all the sympto-
matic cases of the COVID-19 virus in Wuhan [73]; the identification of contagion,
death rates, and recovery of COVID-19-infected individuals in Italy [74]; the
combination of a basic SIR model with probabilistic methods for the estimation of
the number of infected people in France [75]; and the use of the SIR model with a
parameter to maintain social distancing measures, in order to forecast the outbreak
stages of COVID-19 in Brazil [76]. The SIR model is more appropriate for a
localized epidemic outbreak because mostly it only examines temporal dynamics.
Hence, it can be most effective for analysis at a country, city, or state level.
Noteworthy is that the SIR model simplifies COVID-19 data analysis.
Inferentially, the model also assumes that recovered individuals who were infected
can be immune to COVID-19. While unconfirmed reports support this, the model
can capture relevant data that can be viewed to make healthcare decisions in the
control of the COVID-19 pandemic.
6.7.2 Standard logistic regression model

This tool examines the available healthcare data of COVID-19 patients. It aligns
risk factors with certain events like the point of entry, exposure to another infected
person, and the presence of an underlying health condition. The major limitation of
the model is the unavailability of sufficient data or data exclusions that are unac-
counted for [71]. As a result of the high likelihood of selection bias created by these
limitations, COVID-19 risk estimates may be inaccurate. Hence, the standard
logistic regression model may not be suitable for examining the connection
between the medical predictors of a patient and COVID-19-related health events
concerning that patient. If logistic regression must be used in this case, some
patients or some of their healthcare data may be excluded. This may omit the
effects of certain predictors, which would not present a holistic picture of the
situation of the COVID-19 virus in a certain place.
However, the standard logistic regression model is most suitable in predictive
situations to determine one outcome or another [77]. This implies that there can
often always be two possible outcomes. In managing the COVID-19 pandemic, this
may still be useful for clinical trials that generate the presence of the disease or not.
6.7.3 Time-to-event analytics model

This is an alternative to the standard logistic regression model, where the outcome
includes the time of the event occurrence along with details on whether an event
occurred or not. In addition, the time-to-event analysis considers censoring data
that is captured when a patient does not experience effects of the virus at follow-up
times. This analytics model was designed to address time of origin, advance tech-
nique assumptions, and healthcare predictors. Accordingly, estimation techniques
for time-to-event data are developed to use the available medical information on
censored patient data for unbiased survival estimates. From multiple points across
the COVID-19 healthcare for a patient, the techniques can incorporate data to
estimate time ratios, infection rates, and hazard ratios [78].
The models discussed earlier have been used to analyze several COVID-19
data for targeted interventions to mitigate the spread globally. Recent real-time
COVID-19 data from a few European countries are presented and analyzed as a
sample to make health-related and COVID-19 safety decisions.
6.7.4 Results of major real-time analysis

Figure 6.9 is the generated chart from Table 6.2, which highlights the data of
recorded COVID-19 cases in several European countries since its global inception
in March 2021. Figure 6.9 is highly instrumental to clinical experts and policy
makers as the number of cases indicates the rate of spread, what regions or areas are
most vulnerable, and how effective curbing strategies are. These all influence their
decisions about where to invest more resources and what change needs to be made
toward the management of COVID-19. In Table 6.2, the most valuable data are the
absolute change that guides interventions at a given time while considering details
such as the identification of the virus, the number of an infected individual that has
recovered, and the death figures [74]. Hence, the SIR model can be used for its
analysis. First, experts divide the population of a certain area of the region into
groups of those susceptible to getting infected, those already infected, and those
who have recovered or are deceased. Thereafter, data are drawn at each month and
analyzed in relation to data from the previous month. This identifies what is
working, what is not working, and issues that need to be addressed to curb the
spread of the virus. Given the sensitivity of these data, blockchain technology may
be implemented to extract, store, securely transmit and access them [55].
Sample data chart to guide in the control of COVID-19

2,500
2,000
1,500
1,000
500
0
Belgium
Croatia
Cuba
Cyprus
Estonia
Europe
European union
France
Germany
Greece
Hungary
Italy
Lithuania
Monaco
Norway
Romania
Russia
Spain
Sweden
Turkey
Ukraine
United Kingdom
Finland
Iceland
Ireland
Poland
Denmark
March, 2020 July, 2021 Absolute change
Figure 6.9 COVID-19 pandemic data (March 2020 to July 2021)

Table 6.2 Data on COVID-19 cases in European countries from March 2020 to
July 2021
Country March 2020 July 2021 Absolute change Relative change (%)
Belgium 0.01 127.64 127.63 1 063 558
Croatia 0.24 37.65 37.41 15 331
Cuba 0.06 688.72 688.66 10 931 105
Cyprus 4.18 983.42 979.24 23 410
Denmark 0.15 126.62 126.48 85 456
Estonia 0.21 81.2 80.98 37 667
Europe 0.39 172.46 172.07 44 120
European Union 0.62 156.32 155.7 25 195
Finland 0.13 76.36 79.23 61 420
France 0.25 288.81 288.56 115 887
Germany 0.19 19.87 19.68 10 144
Greece 0.1 225.92 255.82 266 479
Hungary 0.13 5.9 5.77 4 336
Iceland 10.88 249.5 238.62 2 192
Ireland 0.17 251.56 251.38 144 474
Italy 3.64 75.52 71.89 1 977
Lithuania 0.05 71.89 71.84 138 156
Monaco 3.64 353.1 349.46 9 601
Norway 0.66 41.9 41.24 6 258
Poland 0.06 2.77 2.71 4 512
Romania 0.02 5.78 5.75 26 159
Russia 0 159.03 159.03 –
Spain 0.25 547.89 547.64 218 185
Sweden 0.18 40.67 40.48 22 002
Turkey 0.03 155.85 1 555.82 519 413
Ukraine <0.01 25.78 25.77 859 100
United Kingdom 0.04 482.1 481.96 346 737
New confirmed cases in Europe (for every 1 million people).
Source: CCSE COVID data, John Hopkins University.
Figures 6.10 and 6.11 represent the results of the analysis of COVID-19 data
recorded in Table 6.3. This accumulated data from 2020 to 2021 was drawn from
the central, eastern, northern, northwestern, and southwestern regions of Virginia,
United States and covers the information of the number of individuals who have
been and not been vaccinated along with the outcomes of their experiences with the
COVID-19 pandemic. Accordingly, these data must contain the personal informa-
tion of individuals in Virginia and should be securely and effectively handled
across the network. These data are valuable to pharmaceuticals and useful in the
several clinical trials involved in developing vaccines that are in global circulation
to fight against the COVID-19 pandemic. Although when infected, living with the
COVID-19 virus and experiencing its symptoms is unavoidable since there is no
proven cure [79], the measure of vaccine effectiveness is evident in the reduced
number of COVID-19 cases, hospitalization, and death rates altogether. Hence, the
Not fully vaccinated COVID-19 cases in regions of Virginia
70,000
60,000
50,000
40,000
30,000
20,000
10,000
0
COVID-19 cases Total hospitalized Total deaths
Central Eastern Northern Northwest Southwest
Figure 6.10 COVID-19 cases and vaccination status in Virginia—not fully

vaccinated
Fully vaccinated COVID-19 cases in regions of Virginia

7,000
6,000
5,000
4,000
3,000
2,000
1,000
0
COVID-19 cases Total hospitalized Total deaths
Central Eastern Northern Northwest Southwest
Figure 6.11 COVID-19 cases and vaccination status in Virginia—fully vaccinated
data in Table 6.3 can be analyzed using the time-to-event analysis, which captures
the effects of the virus experienced by individuals while using estimation techni-
ques to incorporate data time ratio, spread rates, and hazard ratios [78]. With this
model, clinicians and healthcare researchers can interpret outcomes appropriately
to influence the management of the pandemic.
Table 6.3 Data on COVID-19 cases and vaccination status by VDH (August
2021)
Not fully vaccinated (Virginia)
Region Not fully vaccinated COVID-19 cases Total hospitalized Total deaths
Central – 44 306 1 250 622
Eastern – 57 982 2 468 776
Northern – 52 749 1 280 335
Northwest – 37 341 1 088 371
Southwest – 36 047 1 865 643
ALL – 228 425 7 951 2 747
Fully vaccinated (Virginia)
Region Fully vaccinated COVID-19 cases Total hospitalized Total deaths
Central 697 833 713 37 12
Eastern 805 413 699 45 13
Northern 1 427 757 747 28 7
Northwest 632 690 571 33 4
Southwest 543 183 629 75 14
ALL 4 642 482 3 359 218 50
Updated data on COVID-19 cases in Virginia, the United States, from December 2020 to August 2021.
Source: Virginia Department of Health—COVID-19 in Virginia.
Results from Figures 6.10 and 6.11 show some positive evidence in the
effectiveness of the vaccines, as the COVID-19 cases of those that have been fully
vaccinated are more than six times less than reported cases of those who have not
been fully vaccinated all through Virginia. Consequently, hospitalization and the
amount of recorded deaths are lower in those who have been vaccinated. While this
may be enough information to influence clinical trials, other sensitive data such as
the history of an individual’s health conditions and their personal data would be
needed for other control modalities.
Figure 6.12 reflects the updated records of total COVID-19 tests and cases in
selected African countries. This gives an idea about how the pandemic has been
managed across Africa. From Figure 6.12, it is evident that many COVID-19 tests
have been conducted across Africa, and while the cases that emerged are a lot less,
the test numbers are a good move toward the management of COVID-19.
Figure 6.13 also reflects data of the number of individuals who have recovered and
died of the disease. Results generated from Table 6.4 show that the critical cases
from COVID-19 complication may lead to death. While extensive research on the
different variants of COVID-19 is ongoing [80], these data may help clinical
experts consider preventive measures or adopt interventions to manage the disease
better for the purpose of reducing the number of reported COVID-19 cases. Hence,
risk factors and underlying health conditions that increase the susceptibility of an
individual getting infected are important medical information. To analyze these
COVID-19 data of African countries: total tests and total cases

9,000,000
8,000,000
7,000,000
6,000,000
5,000,000
4,000,000
3,000,000
2,000,000
1,000,000
0
Morocco
Tunisia
Ethiopia
Libya
Kenya
Algeria
Nigeria
Mozambique
Botswana
Namibia
Zimbabwe
Ghana
Uganda
Cameroon
Rwanda
Senegal
Malawi
DRC
Angola
Madagascar
Sudan
Mauritania
Egypt
Ivory coast
Total tests Total cases
Figure 6.12 COVID-19 data of African countries showing total tests and cases
COVID-19 data of African countries showing the total cases against

death and recovery records
8,00,000
7,00,000
6,00,000
5,00,000
4,00,000
3,00,000
2,00,000
1,00,000
0
Libya
Morocco
Tunisia
Ethiopia
Kenya
Algeria
Nigeria
Mozambique
Botswana
Namibia
Zimbabwe
Ghana
Uganda
Cameroon
Rwanda
Senegal
Malawi
DRC
Angola
Sudan
Mauritania
Egypt
Ivory coast
Madagascar
Total cases Total deaths Total recovered
Figure 6.13 COVID-19 cases and effects in African countries showing the total
cases vs. death and recovery records (August 2021)
data from Table 6.4, the standard logistic regression model is most suitable, given
that predictive analysis for possible future outcomes is often implemented on these
data. Given that this medical information is shared among health experts managing
a COVID-19 patient, it should be handled securely. Blockchain technology pro-
vides a distributed data handling that addresses this. For a robust blockchain
Table 6.4 Data on COVID-19 cases, death, and recoveries in African countries
S. Country Population Total Total Total Total Active Critical

no. tests cases deaths recov- cases cases
ered
1 Morocco 37 391 929 7 998 148 696 282 10 335 612 390 73 557 1 042
2 Tunisia 11 954 726 2 337 391 610 660 20 931 542 210 47 519 630
3 Egypt 104 454 3 068 679 284 706 16 575 233 227 34 904 90
842
4 Ethiopia 118 087 3 069 384 284 091 4 426 264 617 15 048 291
799
5 Libya 6 974 062 1 364 372 267 846 3 719 199 135 64 992 –
6 Kenya 55 071 577 2 190 806 211 828 4 149 197 307 10 372 133
8 Algeria 44 721 116 230 861 181 376 4 550 121 353 55 473 48
9 Nigeria 211 744 2 542 261 178 086 2 187 165 763 10 136 11
938
10 Mozambique 32 215 423 776 609 132 452 1 613 102 957 27 882 32
11 Botswana 2 403 775 1 525 222 122 574 1 704 104 731 16 139 1
12 Namibia 2 591 466 625 854 121 043 3 191 99 191 18 661 76
13 Zimbabwe 15 100 848 1 149 029 116 327 1 900 88 829 23 598 12
14 Ghana 31 785 183 1 459 845 108 226 874 100 383 6 969 36
15 Uganda 47 323 337 1 499 619 95 723 2 793 90 225 2 705 542
16 Cameroon 27 270 802 1 751 774 82 064 1 334 80 433 297 152
17 Rwanda 13 304 466 2 157 915 76 091 890 44 986 30 215 41
18 Senegal 17 227 840 674 910 67 579 1 482 50 533 15 564 69
19 Malawi 19 672 148 356 144 55 920 1 828 41 155 12 937 285
20 DRC 92 544 772 282 990 51 889 1 048 30 189 20 652 –
21 Ivory Coast 27 094 912 813 541 51 380 341 50 069 970 –
22 Angola 33 986 273 821 071 43 662 1 049 40 122 2 491 8
23 Madagascar 28 471 860 230 721 42 776 948 42 445 – 27
24 Sudan 44 972 864 234 414 37 138 2 776 30 867 3 495 –
25 Mauritania 4 783 697 391 940 28 197 605 23 818 3 774 16
Reported COVID-19 cases and deaths by country.
Source: Worldometer.
technology network, a software development network (SDN) may be combined

with the IoMT that reduces unsecure access to these data and external response
time [81].
The SDN technology also helps health experts identify what countries may
require the investment of more resources, as Figure 6.13 highlights the most vul-
nerable countries. These are countries with the highest number of reported COVID-
19 cases and have recorded the highest number of COVID-related deaths.
6.8 Recommendations and future directions

A discussion of how blockchain technology can be applied to real-time healthcare
data-handling practices is presented as a recommendation and direction for future
practice, to manage COVID-19 and inhibit the spread. These are drawn from the
analysis of COVID-19 data illustrated in the previous section and the limitations of
blockchain technology to be addressed.
First, blockchain technology protects the privacy of patients and authorized
users while preserving data consent and automated provenance in areas such as
healthcare, wherewith strong security, medical information can be shared between
doctors and health facilities; IoMT, where privacy is maintained for the preference
of any IoMT device; and data storage, where there are data protection and data
access control across the network. In-line with this, more blockchain technology
application systems should be built and optimized for better healthcare support that
reduces COVID-19 cases to the lowest minimum and improves other sectors. It has
been established that the potential benefits of blockchain transcend the healthcare
sector to practically all sectors that operate with data and things across a network.
Therefore beyond the COVID-19 pandemic, blockchain technology should be
adopted for better cyber security, data privacy, and communication.
However, limitations such as data policy and regulations, interoperability,
scalability, and response time are acknowledged as the major challenges to be
addressed. For stringent data policies and regulations, extensive research and
healthcare influence may introduce operational standards of blockchain technology
for handling sensitive data like that of the COVID-19 pandemic. Perhaps, only
permissioned blockchain should be adopted for the handling of COVID-19 pan-
demic data to avoid cyberattacks and unauthorized data sharing. Alternatively, the
medical information of patients can be protected by blockchain encryption that
should be updated or maintained regularly. The interoperability still experienced in
some blockchain systems prevents optimal operational communication between
two frameworks, while the synchronization of blockchain with healthcare facilities
persists, particularly in developing countries. This can be addressed with the
introduction of a few subchains that would reduce the pressure on the blockchain
network and allows smaller and quicker transactions on the blockchain. The pro-
cessing speed and efficiency of the system can also be improved with interoper-
ability when fewer nodes are involved in the validation process. However, other
optimization tools should be in place to control the tradeoff.
Scalability also limits the operation of blockchain on the network as it affects
the processing speed and the validation of transactions. It has been established that
larger networks are more costly to run nodes and take more response time. To
address this, using smaller blocks can help to provide a better scale without com-
promising the response time or increasing the processing cost. Smaller blocks are
also compatible with cloud solutions, thereby also providing better interoperability.
Future research direction may be in the design of blocks that provide optimal scale
and interaction within the network.
Existing healthcare applications also experience some security concerns
following the detection of several data leaks and attacks on blockchain applica-
tions across networks. This is critical because valuable data runs blockchain
technology systems. So compromised data defeat one of the main purposes of
adopting blockchain technology to handle data as sensitive as that of the COVID-
19 pandemic. The integration of machine learning and other system security
optimization programs in the blocks are likely to better secure data on the
network.
Given that blockchain technology systems can solve most real-world problems,
future research areas may focus on optimization strategies in areas of privacy,
security, data access, and block size. Furthermore, extensive research on blockchain
technology implementation to improve data handling may be considered since there
are currently only a few ongoing developments of working prototypes that go
through a series of clinical trials and system testing to manage blockchain limitations.
The integration of blockchain technology in big data–processing functions is also yet
to be explored. Therefore, working models for the best data management imple-
mentation practices are needed. This can play a vital role in future healthcare,
communication, IoMT, and identity management systems. However, the acceptance
of these new models would need new data regulations that support blockchain
technologies operations. Nevertheless, blockchain technology systems must comply
with the GDPR in their existing and impending solutions.
Blockchain technology is evolving to revolutionize data handling within and out-

side of healthcare. As seen, it can be directly applied in several applications where
an unalterable record of data is useful. To evaluate blockchain application in
COVID-19 data handling, recent developments, potential benefits, key challenges,
prospects, and the data-handling application of blockchain have been highlighted in
this chapter. Practical solutions for the limitations of implementing blockchain and
productive research focus were also identified. Overall, integrating optimization
programs in the blockchain to help expand its privacy, security, authenticating,
interoperability and scalability functions were recommended. For implementations
in other sectors where seamless data management is vital, operational standards for
handling data effectively may be introduced for blockchain implementation.
Although some working prototypes to improve, the functions of blockchain tech-
nology have been developed, while others like permissible blockchain, off-chain
storage, and smart contracts have been tested in many real-world applications.
Nevertheless, more research is needed for optimization and evaluation of block-
chain implementation and supporting ongoing efforts. This is aimed at addressing
issues with blockchain technology use such as interoperability, response delay,
security, scalability, and privacy.
For healthcare data handling, as in the case of the COVID-19 global pandemic,
blockchain technology is increasingly implemented to create a reasonable basis for
an effective evidence-based treatment or intervention process. Existing research
has shown that a blockchain is an effective tool for safe data storage and sharing
between authorized users. It can also be used to improve the protocols of risk
management. It can then be concluded that blockchain technology is highly
instrumental in the clinical practice of controlling COVID-19. However, a few
issues that have been experienced in the pandemic are data manipulation,
miscommunication, delayed response time, and failure at single points. This study
has also shown that blockchain can be used for other healthcare functions like
pharmaceutical supply-chain management, counterfeit drug detection, remote
health monitoring, and biomedical research. Therefore, it is important to regularly
update or maintain these systems for optimal functions in controlling the COVID-
19 and tracking other infectious diseases. As evidence, key elements of the
blockchain, like already existing smart contracts, may be expanded to enable easy
interaction between users and the blockchain, while blockchain applications may
be extended to accommodate a larger amount of data.
Acknowledgments
The work of Agbotiname Lucky Imoize is supported by the Nigerian Petroleum

Technology Development Fund (PTDF) and the German Academic Exchange
Service (DAAD) through the Nigerian-German Post-graduate Program under grant
57473408.
References
[1] A. Azim, M. N. Islam, and P. E. Spranger, “Blockchain and novel cor-
onavirus: Towards preventing COVID-19 and future pandemics,” Iberoam.
J. Med., vol. 03, pp. 215–218, 2020, [Online]. Available: http://creative-
commons.org/licenses/by/4.0/.
[2] N. T. Lee, J. Karsten, and J. Roberts, “Removing regulatory barriers to tel-
ehealth before and after COVID-19,” Brookings Inst., pp. 1–24, 2020,
[Online]. Available: https://www.brookings.edu/wp-content/uploads/2020/
05/Removing-barriers-to-telehealth-before-and-after-COVID-19_PDF.pdf.
[3] D. C. Nguyen, M. Ding, P. N. Pathirana, and A. Seneviratne, “Blockchain
and AI-based solutions to combat Coronavirus (COVID-19)-like epidemics:
A survey,” IEEE Access, vol. 9, pp. 95730–95753, 2021.
[4] M. Hiwale, S. Phanasalkar, and K. Kotecha, “Using blockchain and dis-
tributed machine learning to manage decentralized but trustworthy disease
data,” Sci. Technol. Libr., vol. 40, no. 2, pp. 190–213, 2021, doi: 10.1080/
0194262X.2020.1859046.
[5] A. Hasselgren, K. Kralevska, D. Gligoroski, S. A. Pedersen, and A. Faxvaag,
“Blockchain in healthcare and health sciences—A scoping review,” Int. J.
Med. Inform., vol. 134, p. 104040, 2020, doi: 10.1016/j.ijmedinf.2019.104040.
[6] C. Agbo, Q. Mahmoud, and J. Eklund, “Blockchain technology in health-
care: A systematic review,” Healthcare, vol. 7, no. 2, p. 56, 2019, doi:
10.3390/healthcare7020056.
[7] A. F. da Conceição, F. S. C. da Silva, V. Rocha, A. Locoro, and J. M.
Barguil, “Electronic health records using blockchain technology,” 2018,
[Online]. Available: http://arxiv.org/abs/1804.10078.
[8] P. V. Kakarlapudi and Q. H. Mahmoud, “A systematic review of blockchain

for consent management,” Healthcare, vol. 9, no. 2, 2021, doi: 10.3390/
healthcare9020137.
[9] R. Wasim, K. Salah, R. Jayaraman, I. Yaqoob, S. Ellahham, and M. Omar,
“Since January 2020 Elsevier has created a COVID-19 resource centre with
free information in English and Mandarin on the novel coronavirus COVID-
19. The COVID-19 resource centre is hosted on Elsevier Connect, the
company’s public news and information,” January 2020.
[10] I. Yaqoob, K. Salah, R. Jayaraman, and Y. Al-Hammadi, “Blockchain for
healthcare data management: Opportunities, challenges, and future recom-
mendations,” Neural Comput. Appl., 2021, doi: 10.1007/s00521-020-05519-w.
[11] M. S. Ali, M. Vecchio, M. Pincheira, K. Dolui, F. Antonelli, and M. H.
Rehmani, “Applications of blockchains in the Internet of Things: A com-
prehensive survey,” IEEE Commun. Surv. Tutorials, vol. 21, no. 2, pp. 1676–
1717, 2019, doi: 10.1109/COMST.2018.2886932.
[12] P. P. Ray, Di. Dash, K. Salah, and N. Kumar, “Blockchain for IoT-based
healthcare: Background, consensus, platforms, and use cases,” IEEE Syst. J.,
vol. 15, no. 1, pp. 85–94, 2021, doi: 10.1109/JSYST.2020.2963840.
[13] M. G. R. Alam, S. F. Abedin, S. I. Moon, A. Talukder, and C. S. Hong,
“Healthcare IoT-Based Affective State Mining Using a Deep Convolutional
Neural Network,” IEEE Access, vol. 7, pp. 75189–75202, 2019.
[14] F. Ali, S. El-Sappagh, S. M. Riazul Islam, et al., “A smart healthcare mon-
itoring system for heart disease prediction based on ensemble deep learning
and feature fusion,” Inf. Fusion, vol. 63, pp. 208–222, 2020, doi: 10.1016/j.
inffus.2020.06.008.
[15] H. Tao, M. Z. A. Bhuiyan, A. N. Abdalla, M. M. Hassan, J. M. Zain, and T.
Hayajneh, “Secured data collection with hardware-based ciphers for IoT-
based healthcare,” IEEE Internet Things J., vol. 6, no. 1, pp. 410–420, 2019,
doi: 10.1109/JIOT.2018.2854714.
[16] C. Chakraborty and M. Rathi, “Smart healthcare systems using big data,” In
Demystifying Big Data, Machine Learning, and Deep Learning for
Healthcare Analytics, (pp. 17–32). London: Academic Press; 2021.
[17] V. Chamola, V. Hassija, V. Gupta, and M. Guizani, “A comprehensive
review of the COVID-19 pandemic and the role of IoT, drones, AI, block-
chain, and 5G in managing its impact,” IEEE Access, vol. 8, pp. 90225–
90265, 2020, doi: 10.1109/ACCESS.2020.2992341.
[18] W. Chen, Z. Xu, S. Shi, Y. Zhao, and J. Zhao, “A survey of blockchain
applications in different domains,” ACM Int. Conf. Proceeding Ser., pp. 17–
21, 2018, doi: 10.1145/3301403.3301407.
[19] D. B. Rawat, V. Chaudhary, and R. Doku, “Blockchain technology:
Emerging applications and use cases for secure and trustworthy smart sys-
tems,” J. Cybersecur. Priv., vol. 1, no. 1, pp. 4–18, 2020, doi: 10.3390/
jcp1010002.
[20] M. C. Suciu, C. Năsulea, and D. Năsulea, “Is blockchain a new creative

industry?,” Collab. Res. Excell. Econ. Soc. Sci., pp. 154–163, 2021, doi:
10.2478/9788366675322-019.
[21] M. Gaynor, J. Tuttle-Newhall, J. Parker, A. Patel, and C. Tang, “Adoption of
blockchain in health care,” J. Med. Internet Res., vol. 22, no. 9, pp. 1–10,
2020, doi: 10.2196/17423.
[22] S. Dash, C. Chakraborty, S. K. Giri, S. K. Pani, and J. Frnda, “BIFM: Big-
data driven intelligent forecasting model for COVID-19,” IEEE Access,
vol. 9, pp. 97505–97517, 2021, doi: 10.1109/ACCESS.2021.3094658.
[23] Q. Li, H. Lin, X. Wei, et al. “MaraVis: Representation and coordinated
intervention of medical encounters in urban marathon,” In Proceedings of
the 2020 CHI Conference on Human Factors in Computing Systems, (pp. 1–
12). New York, NY: ACM; 2020.
[24] X. Yue, H. Wang, D. Jin, M. Li, and W. Jiang, “Healthcare data gateways:
Found healthcare intelligence on blockchain with novel privacy risk
control,” J. Med. Syst., vol. 40, no. 10, 2016, doi: 10.1007/s10916-016-
0574-6.
[25] A. H. Mayer, C. A. da Costa, and R. da Rosa Righi, “Electronic health
records in a Blockchain: A systematic review,” Health Informatics J.,
vol. 26, no. 2, pp. 1273–1288, 2020, doi: 10.1177/1460458219866350.
[26] S. Dash, C. Chakraborty, S. K. Giri, and S. K. Pani, “Intelligent computing
on time-series data analysis and prediction of COVID-19 pandemics”,
Pattern Recognit. Lett., vol. 151, pp. 69–75, 2021, https://doi.org/10.1016/j.
patrec.2021.07.027.
[27] O. Gutiérrez, G. Romero, L. Pérez, A. Salazar, P. Wightman, and M. Charris,
“HealthyBlock: Blockchain-based it architecture for electronic medical
records resilient to connectivity failures,” Int. J. Environ. Res. Public Health,
vol. 17, no. 19, pp. 1–38, 2020, doi: 10.3390/ijerph17197132.
[28] M. M. Qureshi, A. Farooq, and M. Mazhar, “Current eHealth challenges and
recent trends in eHealth applications,” February 2021.
[29] B. J. Nievas Soriano, S. Garcia Duarte, A. M. Fernandez Alonso, A. Bonillo
Perales, and T. Parron Carreno, “eHealth: Advantages, disadvantages and
guiding principles for the future,” J. Med. Internet Res., pp. 1–23, 2019, doi:
10.2196/preprints.15366.
[30] P. Ratta, A. Kaur, S. Sharma, M. Shabaz, and G. Dhiman, “Application of
blockchain and Internet of Things in healthcare and medical sector:
Applications, challenges, and future perspectives,” J. Food Qual., vol. 2021,
2021, doi: 10.1155/2021/7608296.
[31] S. Shi, D. He, L. Li, N. Kumar, and M. Khurram, “Since January 2020
Elsevier has created a COVID-19 resource centre with free information in
English and Mandarin on the novel coronavirus COVID-19. The COVID-19
resource centre is hosted on Elsevier Connect, the company’s public news
and information,” January 2020.
[32] A. Grech, I. Sood, and L. Ariño, “Blockchain, self-sovereign identity and

digital credentials: Promise versus praxis in education,” Front. Blockchain,
vol. 4, pp. 1–12, 2021, doi: 10.3389/fbloc.2021.616779.
[33] S. Parvathavarthini, N. Karthikeyani Visalakshi, S. Shanthi, and J. Madhan
Mohan, “An improved crow search based intuitionistic fuzzy clustering
algorithm for healthcare applications,” Intell. Autom. Soft Comput., vol. 26,
no. 2, pp. 253–260, 2020, doi: 10.31209/2019.100000155.
[34] W. Bible, J. Raphael, M. Riviello, P. Taylor, and I. O. Valiente, “Blockchain
technology and its potential impact on the audit and assurance profession,”
CPA Canada, AICPA, pp. 1–28, 2017, [Online]. Available: https://www.
aicpa.org/content/dam/aicpa/interestareas/frc/assuranceadvisoryservices/
downloadabledocuments/blockchain-technology-and-its-potential-impact-on-
the-audit-and-assurance-profession.pdf.
[35] Institute of Medicine. Patient Safety: Achieving a New Standard for Care.
Washington, DC: National Academies Press; 2004. https://doi.org/10.17226/
10863.
[36] M. L. Jochumsen and A. Chaudhuri, “Blockchain’s impact on supply chain
of a pharmaceutical company,” EUROMA Conf. 2018, pp. 1–8, June 2018.
[37] M. Sahoo, S. S. Singhar, and S. S. Sahoo, “A blockchain based model to
eliminate drug counterfeiting,” Adv. Intell. Syst. Comput., vol. 1101,
pp. 213–222, 2020, doi: 10.1007/978-981-15-1884-3_20.
[38] A. Sanghi, Aayush, A. Katakwar, A. Arora, and A. Kaushik, “Detecting fake
drugs using blockchain,” Int. J. Recent Technol. Eng., vol. 10, no. 1,
pp. 100–109, 2021, doi: 10.35940/ijrte.a5744.0510121.
[39] H. K. Bhuyan, C. Chakraborty, Y. Shelke, and S. K. Pani, “COVID-19
diagnosis system by deep learning approaches,” Expert Syst., pp. 1–18, 2021,
doi: 10.1111/exsy.12776.
[40] S. G. Alonso, J. Arambarri, M. López-Coronado, and I. de la Torre Dı́ez,
“Proposing New Blockchain Challenges in eHealth,” J. Med. Syst., vol. 43,
no. 3, 2019, doi: 10.1007/s10916-019-1195-7.
[41] A. A. Vazirani, O. O’Donoghue, D. Brindley, and E. Meinert,
“Implementing blockchains for efficient health care: Systematic review,” J.
Med. Internet Res., vol. 21, no. 2, pp. 1–12, 2019, doi: 10.2196/12439.
[42] P. E. Velmovitsky, F. M. Bublitz, L. X. Fadrique, and P. P. Morita,
“Blockchain applications in health care and public health: Increased trans-
parency,” JMIR Med. Inform., vol. 9, no. 6, 2021, doi: 10.2196/20713.
[43] A. Fusco, G. Dicuonzo, V. Dell’atti, and M. Tatullo, “Blockchain in
healthcare: Insights on COVID-19,” Int. J. Environ. Res. Public Health,
vol. 17, no. 19, pp. 1–12, 2020, doi: 10.3390/ijerph17197167.
[44] R. S. Evans, “Electronic health records: Then, now, and in the future,”
Yearb. Med. Inform., pp. S48–S61, 2016, doi: 10.15265/IYS-2016-s006.
[45] A. Sheikh, M. Anderson, S. Albala, et al., “Health information technology
and digital innovation for national learning health and care systems,” The
Lancet Digital Health, vol. 3, no. 6, pp. e383–e396, 2021, doi: 10.1016/
S2589-7500(21)00005-4.
[46] D. Tith, J.-S. Lee, H. Suzuki, et al., “Application of blockchain to main-

taining patient records in electronic health record for enhanced privacy,
scalability, and availability,” Healthc. Inform. Res., vol. 26, no. 1, pp. 3–12,
2020, doi: 10.4258/hir.2020.26.1.3.
[47] J. Taskinsoy, “Blockchain: A misunderstood digital revolution. Things you
need to know about blockchain,” SSRN Electron. J., 2019, doi: 10.2139/
ssrn.3466480.
[48] S. Halilbegovic and A. O. Arapovic, “Impact of blockchain technology on
the payment management systems – What future holds?,” Eur. J. Econ.
Stud., vol. 8, no. 1, 2019, doi: 10.13187/es.2019.1.43.
[49] H. F. Atlam, M. A. Azad, A. G. Alzahrani, and G. Wills, “A review of
blockchain in Internet of Things and AI,” Big Data Cogn. Comput., vol. 4,
no. 4, pp. 1–27, 2020, doi: 10.3390/bdcc4040028.
[50] M. Nicola, C. Sohrabi, G. Mathew et al., “Health policy and leadership
models during the COVID-19 pandemic: A review,” Int. J. Surg., vol. 81,
pp. 122–129, 2020, [Online]. Available: /pmc/articles/PMC7366988/%0A/
pmc/articles/PMC7366988/?report¼abstract%0Ahttps://www.ncbi.nlm.nih.
gov/pmc/articles/PMC7366988/.
[51] M. N. Kamel Boulos, J. T. Wilson, and K. A. Clauson, “Geospatial block-
chain: Promises, challenges, and scenarios in health and healthcare,” Int. J.
Health Geogr., vol. 17, no. 1, pp. 1–10, 2018, doi: 10.1186/s12942-018-
0144-x.
[52] I. Radanović and R. Likić, “Opportunities for use of blockchain technology
in medicine,” Appl. Health Econ. Health Policy, vol. 16, no. 5, pp. 583–590,
2018, doi: 10.1007/s40258-018-0412-8.
[53] J. Yli-Huumo, D. Ko, S. Choi, S. Park, and K. Smolander, “Where is current
research on Blockchain technology? – A systematic review,” PLoS One,
vol. 11, no. 10, pp. 1–27, 2016, doi: 10.1371/journal.pone.0163477.
[54] D. Marbouh, T. Abbasi, F. Maasmi, et al., “Blockchain for COVID-19:
Review, opportunities, and a trusted tracking system,” Arab. J. Sci. Eng.,
vol. 45, pp. 9895–9911, 2020, doi: 10.1007/s13369-020-04950-4.
[55] Z. Zheng, S. Xie, H. Dai, X. Chen, and H. Wang, “An overview of block-
chain technology: Architecture, consensus, and future trends,” Proc. – 2017
IEEE 6th Int. Congr. Big Data, BigData Congr. 2017, pp. 557–564, July
2017, doi: 10.1109/BigDataCongress.2017.85.
[56] Q. Xia, E. B. Sifah, A. Smahi, S. Amofa, and X. Zhang, “BBDS:
Blockchain-based data sharing for electronic medical records in cloud
environments,” Information, vol. 8, no. 2, 2017, doi: 10.3390/info
8020044.
[57] W. J. Gordon and C. Catalini, “Blockchain technology for healthcare:
Facilitating the transition to patient-driven interoperability,” Comput. Struct.
Biotechnol. J., vol. 16, pp. 224–230, 2018, doi: 10.1016/j.csbj.2018.06.003.
[58] E. Pritchard, P. C. Matthews, N. Stoesser, et al., “Impact of vaccination on
new SARS-CoV-2 infections in the United Kingdom,” Nat. Med., vol. 27,
pp. 1370–1378, 2021, doi: 10.1038/s41591-021-01410-w.
[59] S. T. Argaw, J. R. Troncoso-Pastoriza, D. Lacey, et al., “Cybersecurity of

hospitals: Discussing the challenges and working towards mitigating the
risks,” BMC Med. Inform. Decis. Mak., vol. 20, no. 1, pp. 1–10, 2020, doi:
10.1186/s12911-020-01161-7.
[60] T. K. Mackey, K. Miyachi, D. Fung, S. Qian, and J. Short, “Combating
health care fraud and abuse: Conceptualization and prototyping study of a
blockchain antifraud framework,” J. Med. Internet Res., vol. 22, no. 9, pp. 1–
24, 2020, doi: 10.2196/18623.
[61] P. Pandey and R. Litoriya, “Securing E-health networks from counterfeit
medicine penetration using blockchain,” Wirel. Pers. Commun., vol. 117, no.
1, pp. 7–25, 2021, doi: 10.1007/s11277-020-07041-7.
[62] H. I. Miller and W. Winegarden, “Fraud in your pill bottle the
unacceptable cost of counterfeit medicines,” October 2020, [Online].
Available: https://www.pacificresearch.org/wp-content/uploads/2020/10/
CounterfeitMed_F.pdf.
[63] S. Tanwar, K. Parekh, and R. Evans, “Blockchain-based electronic health-
care record system for healthcare 4.0 applications,” J. Inf. Secur. Appl.,
vol. 50, 2020, doi: 10.1016/j.jisa.2019.102407.
[64] P. Pandey and R. Litoriya, “Implementing healthcare services on a large
scale: Challenges and remedies based on blockchain technology,” Heal.
Policy Technol., vol. 9, no. 1, pp. 69–78, 2020, doi: 10.1016/j.
hlpt.2020.01.004.
[65] I. A. Omar, R. Jayaraman, K. Salah, and M. C. E. Simsekler, “Exploiting
Ethereum smart contracts for clinical trial management,” Proc. IEEE/ACS
Int. Conf. Comput. Syst. Appl. AICCSA, vol. 2019, August 2019, doi:
10.1109/AICCSA47632.2019.9035341.
[66] F. Carlini, R. Carlini, S. Dalla Palma, R. Pareschi, and F. Zappone, “The
Genesy model for a blockchain-based fair ecosystem of genomic data,”
Front. Blockchain, vol. 3, pp. 1–13, 2020, doi: 10.3389/fbloc.2020.483227.
[67] H. Szajewska, “Evidence-based medicine and clinical research: Both are
needed, neither is perfect,” Ann. Nutr. Metab., vol. 72, no. Suppl. 3, pp. 13–
23, 2018, doi: 10.1159/000487375.
[68] S. Tian, W. Yang, J. M. Le Grange, P. Wang, W. Huang, and Z. Ye, “Smart
healthcare: Making medical care more intelligent,” J. Global Health, vol. 3,
no. 3, pp. 62–65, 2019, doi: 10.1016/j.glohj.2019.07.001.
[69] N. G. Badr, “Blockchain or distributed ledger technology what is in it for the
healthcare industry?,” IC3K 2019 – Proc. 11th Int. Jt. Conf. Knowl. Discov.
Knowl. Eng. Knowl. Manag., vol. 3, pp. 277–284, 2019, doi: 10.5220/
0008348902770284.
[70] Deloitte, “Blockchain—A New Model for Health Information Exchanges,”
pp. 1–12, 2016. Available from: https://www.healthit.gov/sites/default/files/
4-37-hhs_blockchain_challenge_deloitte_consulting_llp.pdf.
[71] M. Wolkewitz, J. Lambert, M. von Cube, et al., “Statistical analysis of
clinical covid-19 data: A concise overview of lessons learned, common
errors and how to avoid them,” Clin. Epidemiol., vol. 12, pp. 925–928, 2020,
doi: 10.2147/CLEP.S256735.
[72] J. Chu, “A statistical analysis of the novel coronavirus (COVID-19) in Italy
and Spain,” PLoS One, vol. 16, no. 3, p. e0249037, 2021.
[73] J. T. Wu, K. Leung, M. Bushman, et al., “Estimating clinical severity of
COVID-19 from the transmission dynamics in Wuhan, China,” Nat. Med.,
vol. 26, no. 4, pp. 506–510, 2020, doi: 10.1038/s41591-020-0822-7.
[74] G. C. Calafiore, C. Novara, and C. Possieri, “A Modified SIR Model for the
COVID-19 Contagion in Italy,” Proc. IEEE Conf. Decis. Control, vol. 2020-
Decem, pp. 3889–3894, 2020, doi: 10.1109/CDC42340.2020.9304142.
[75] L. Roques, E. K. Klein, J. Papaı̈x, A. Sar, and S. Soubeyrand, “Using early
data to estimate the actual infection fatality ratio from covid-19 in France,”
Biology (Basel), vol. 9, no. 5, pp. 1–11, 2020, doi: 10.3390/biology9050097.
[76] S. B. Bastos and D. O. Cajueiro, “Modeling and forecasting the early evo-
lution of the Covid-19 pandemic in Brazil,” Sci. Rep., vol. 10, no. 1, pp. 1–
15, 2020, doi: 10.1038/s41598-020-76257-1.
[77] M. E. Shipe, S. A. Deppen, F. Farjah, and E. L. Grogan, “Developing pre-
diction models for clinical use using logistic regression: An overview,” J.
Thorac. Dis., vol. 11, no. Suppl. 4, pp. S574–S584, 2019, doi: 10.21037/
jtd.2019.01.25.
[78] J. Le-Rademacher and X. Wang, “A Time-to-event data: An overview and
analysis considerations,” J. Thorac. Oncol., vol. 16, no. 7, pp. 1067–1074,
2021. doi: 10.1016/j.jtho.2021.04.004.
[79] L. J. Muhammad, E. A. Algehyne, S. S. Usman, A. Ahmad, C. Chakraborty,
and I. A. Mohammed, “Supervised machine learning models for prediction
of COVID-19 infection using epidemiology dataset,” SN Comput. Sci.,
vol. 2, no. 1, 2021, doi: 10.1007/s42979-020-00394-7.
[80] A. S. Lauring and E. B. Hodcroft, “Genetic variants of SARS-CoV-2 – What
do they mean?,” JAMA – J. Am. Med. Assoc., vol. 325, no. 6, pp. 529–531,
2021, doi: 10.1001/jama.2020.27124.
[81] A. Rahman, C. Chakraborty, A. Anwar, et al., “SDN–IoT empowered intel-
ligent framework for industry 4.0 applications during COVID-19 pandemic,”
Cluster Comput., pp. 1–18, 2021, doi: 10.1007/s10586-021-03367-4.
Chapter 7
Social distancing technologies for COVID-19
Aumnat Tongkaw1
Abstract
Because of the COVID-19 scenario, the viral infection rate in Thailand has sky-
rocketed as people commute to work and go about their everyday lives. Social
divides have a role in illness prevention. As a result, Thailand is focusing on
adopting different technologies to assist us in managing the gap and support
everyone’s ability to work. Adopting this technology has resulted in it being the
New Normal in use as a substitute for older systems. It also plays a role in instilling
everyone a sense of social responsibility and caution to avoid near individuals from
becoming infected; it has become an accepted part of daily life. This chapter offers
a template analysis of stakeholder interviews conducted with the aid of technology
to accomplish social distancing, which is divided into three categories: educational,
public health, and social distancing, as well as manufacturing employees. A soft-
ware usage pattern has been developed due to data analysis utilizing the content
analysis to help create social distancing for job planning and everyday use. It can
also be used to plan future smart cities.
Keywords: COVID-19; Social distancing; Healthcare; Education; Manufacturing
7.1 Introduction
Since the spread of the COVID-19 pandemic, the government in several nations has
adopted the phrase “social distancing.” Social distancing is a physical distancing
technique that involves establishing a distance between two individuals or among
people to limit their interaction with one another. This scenario has a considerable
influence on how the population work and interact with one another. The accept-
ability of the distance is no more than 6 ft apart. In public health, social distancing
is frequently referred to as physical distancing. In a collection of non-
pharmaceutical measures aimed at preventing the transmission of a contagious
illness by keeping a physical barrier between individuals, minimizing the number
1
Faculty of Science and Technology, Songkhla Rajabhat University, Songkhla, Thailand
of times people came into contact with one another, and limiting huge crowd
gatherings. Scientists and engineers are using all available technologies to combat
the virus’s spread; ICT can provide valuable tools for ensuring social distancing. To
be a generation in which technology has improved hardware and software,
including mobile applications and many Internet of Things (IoT) sensor actuators in
devices. Artificial intelligence (AI) techniques such as machine learning, data
mining, deep learning, and other AI approaches are also included in some tech-
nologies. For 2019-nCoV pandemic patients, these options would make diagnosis
and prognosis easier. In one study, for example, supervised machine learning
models for COVID-19 infection were developed [1]. People are using social media
more than ever before during COVID-19. Technological developments and social
media platforms give opportunities to keep people safe, informed, and connected.
Working from home has become possible because to access to social media during
COVID-19. Furthermore, technology enables people to interact online or work
from home in order to preserve social distance. Every firm in Thailand, both gov-
ernment and commercial, has embraced social media and technology to improve
work efficiency even further.
7.2 Methodology
The content of this chapter came from the 30 participants in three sectors: education,
industry, and healthcare. The literature draws the theme of interview questions and
analysis by using template analysis that is a part of thematic analysis [2,3]. Template
emphasizes hierarchical coding while balancing a high level of organization in the
process of analyzing textual data with the flexibility to adjust it to the needs of a
specific study [3]. Researchers questioned 30 persons involved in utilizing technology
to assist in times of need social distance, interpreting the details from the interview,
recording it, and documenting it. Subsequently, preliminary coding was performed for
important information the interviewees had discussed the same issue or are essential
or a focus, which may lead to the main point of using those technologies. The
researchers recorded the respondents’ remarks regarding problems of interest and
impediments to leadership. Following that, the researcher gathered several issues—
both primary and sub-issues—which were discussed, and the intricacies of each issue
were discovered in depth. The researchers then published this article based on the
resultant template to demonstrate to readers the uniformity of technology imple-
mentations utilized in three areas: education, public health, and manufacturing.
7.3 Social distancing technologies for education

During the COVID-19 epidemic, many institutes must teach entirely online at all
educational levels, from preschool to university. As a result, the ministry of edu-
cation employs social distancing to employ social distancing between individuals,
which significantly influences learning activities at higher institutions. Colleges
replace in-person meetings with online networking and homework obligations. This
Social distancing technologies for COVID-19 183
study attempts to characterize the various models and platforms utilized in online
learning for future instructors during the social distance phase [4]. Many online
teaching and learning tools are proliferating. Some university courses that need
practice laboratories must consider how to isolate classes of students and how to
alert when students are close to one other in the lab. As a result of the COVID-19
epidemic, many nations have had to adjust to new conditions in various fields,
including education. The Indonesian government has decided to shift education
away from face-to-face meetings and video conferencing using various learning
management systems (LMSs) such as Moodle and Google Classroom.
7.3.1 Learning management system

During COVID-19, LMSs made a significant contribution. Furthermore, keeping
all of the course information, videos, and activities in one location might help
manage the course and stick to the lesson schedule. Students may access them
anytime, from anywhere in the globe, with security remaining the primary concern.
Students can access the Internet using suitable devices such as computers, tablets,
or even mobile phones. Students can utilize LMS to learn while schools are closed
due to social distancing. Furthermore, school teachers can reduce the administrative
burdens involved with keeping instructional resources in their classrooms, saving
educational institution’s costs. The benefits of the LMS system include the ability
for schools to administer the system themselves, alter the system to suit the learners
and the environment of each school without too much effort, and the fact that it
does not require as many computer specialists as other systems. The LMS is
designed to enable integrated teaching techniques, which combine self-paced
eLearning with instructor-led online instruction. Teachers may utilize the inte-
grated authoring tool to import current courses or build completely new ones from
the start and then establish adjustable student paths that determine when and how
learners can take them. Discussion forums and private chats increase class parti-
cipation, and teachers may plan live webinars to teach students real time.
Automated grading streamlines homework assignment evaluations, while detailed
data and student analytics aid professors in evaluating which students need more
help and which courses may be improved. Many countries must adjust to changing
conditions in a variety of sectors, including education. As a consequence of the
COVID-19 epidemic, LMS is being utilized in Thailand education for teaching and
learning, contributing to teacher self-improvement. The LMS system can also plan
education until the conclusion of the semester online. As a result, research has
proven that a completely online education system is feasible [5]. Indonesia’s gov-
ernment has held online meetings and has made all schooling available online.
They make use of platforms like massive open online course (MOOC) Moodle and
Google Classroom [6].
7.3.1.1 Massive open online course

MOOCs have emerged as one of Thailand’s most effective online learning tech-
niques following COVID-19. The Thai MOOC platform includes a large and
approved library of educational content that may be accessed in various ways [7].
Su
Figure 7.1 Thai MOOC platform
Following COVID-19, MOOCs emerged as one of Thailand’s most effective online

learning platforms. The Thai MOOC platform provides a vast and approved library
of educational resources that may be accessed in various ways. See Figure 7.1 that
shows Thai MOOC platform. The effectiveness of the MOOC on the learners
depends on the learn–certify–deploy ideology. The quick process helps people
integrate their knowledge gained from the class to help them do their job better or
follow their interests. Interestingly, Thais people who take MOOC courses tend to
follow their interests rather than other goals. Fast implementation is an important
factor contributing to MOOCs’ success during the pandemic. Most people stepped
out of their normal lives before and exerted their willingness to learn something
new [8].
Lessons obtained through designing instructional technology and guiding
articles, on the other hand, may be implemented in a variety of situations in primary
care education. It is discovered that MOOCs provided the freedom to adjust the
content, modify the instructional phases, and adjust the components based on the
pandemic’s progress as well as learners’ requirements and suggestions.
Furthermore, MOOCs are ideal for rapid information distribution. Learner partici-
pation in virtual group discussions was advantageous and was widely recognized
by users as an effective tool to share insights globally [9].
7.3.1.2 Google Classroom

Google Classroom is a free online service for schools, nonprofits, and anyone with
a Google account. Google Classroom makes it easy for students and teachers to stay
connected, both inside and outside the classroom. It is a hybrid learning platform
originating from Google that promises to facilitate the creation, distribution, and
assignment of assignments in a paperless way. Using this teachers can create virtual
classes, invite students to join classes, and provide information about the teaching
and learning process. It can also provide teaching materials that students can learn
in the form of material presentation files and learning videos, give assignments to
Figure 7.2 Google Classroom
students, and make a schedule for collecting assignments and others. Being a
relatively new platform for learning, the need for Google Classroom in online
Indonesian learning is worth investigating from various perspectives, both from the
side of students, teachers, and supporting facilities. For example, from the student
aspect, active student involvement, accuracy in doing assignments, student enthu-
siasm, and student learning outcomes. From the teacher’s perspective, for example,
the teacher’s mastery of information technology, the teacher’s skills in preparing
lesson plans, implementing learning, and providing evaluations. Meanwhile, in
terms of infrastructure, the examples are the availability of a strong network,
Internet connection, and the financial readiness of parents and perhaps teachers to
purchase data packages. This Best Practice will not reveal all aspects related to the
implementation of online learning using Google Classroom during the COVID-19
pandemic but is only limited to increasing student activity and learning outcomes
through online learning using Google Classroom [10]. Google Classroom is a free
application that allows teachers to create courses, give and mark homework, and
offer feedback to students. Instructors may set permissions for students, parents,
and co-teachers, share their display, connect a student’s screen and lockdown for a
pop quiz. Teachers can also employ digital learning to augment classroom educa-
tion. Figure 7.2 shows many classes in a Google Classroom in a university of
Thailand.
7.3.1.3 Moodle
Moodle is a program that manages the LMS, or for short, the LMS. That is the LMS
Martin Dougiamas created for use in online teaching and learning. The system
management structure is a Web-Based Instruction that has a vital component in the
writing language, which includes PHP and MySQL database, which provides
Figure 7.3 Moodle platform in a university of Thailand
administrators, instructors, and students, the three user groups. Interaction is

prioritized in network-based teaching and learning. Moodle is open-source soft-
ware, which indicates that it is free to use. The license is a GPL (General Public
License) that users can download. The program is free to use and can develop
additional programs according to users.
Moodle-based LMS is effective for improving students’ argumentation skills.
It is supported by the statement of many researches which state that learning using
LMS can be used to train students’ thinking skills in various courses. Furthermore,
based on the results of [11], it demonstrates that the usage of Moodle in blended
learning has an average degree of effectiveness in terms of student motivation, with
a mean of 3.216, a mean of 3.164 in terms of student success, and a mean of
3.199 in terms of student interaction and coordination. Several theoretical reasons
can be used to justify that the student’s argumentation ability has moderate
improvement criteria. Moodle and its procedures are associated with the applica-
tion of learning through the use of a Moodle-based LMS. A well-designed Moodle-
based LMS will facilitate learning by leveraging self-service and self-guided ser-
vices, quickly collecting and distributing learning information, and facilitating
knowledge reuse. This idea allows teachers to manage classwork and communicate
with students in a flexible and timely manner. Figure 7.3 shows an own Moodle-
based in a university in Thailand.
7.3.2 Social networking and conference software for

education
In the Thai education system, there are two levels of LMS: school and under-
graduate and above. In Thailand, Google Classroom is largely utilized. However,
there is a difficulty in that most instructors cannot use Google Classroom since
pupils do not have personal computers. Then, the usage of LMS is restricted, as it
does not cover all courses. During the lockdown, Thai schools allow pupils to study
online using various technologies. According to the interview data, the most
commonly utilized system is a social media platform that parents or kids use on a
daily basis to contact friends or teachers at school, such as LINE or Facebook.
Students utilize social media to connect with one another and form groups for
teaching and learning. Students also utilize conference software for communicating
and teaching, such as Google Meet, Zoom, Webex, Team, or LINE Group call.
7.4 Social distancing technology in healthcare
Many patients require hospitalization, particularly in underdeveloped countries.

The following examples demonstrate how technology may help with social dis-
tancing policies by lowering the number of people queuing at the hospital, waiting
for prescriptions, or completing payments.
7.4.1 Wearable technology

Wearable technology is becoming more prevalent in our daily life as well as the
healthcare profession. These regions can benefit from on-time remote access,
symptom prognosis, and interaction tracking enabled by technology. The IoT plays
an important part in tracking a person’s health and diagnosing COVID-19. ML is a
sophisticated AI technique that can be used to detect and diagnose COVID-19 from
X-ray and CT pictures reliably, and it could be a possible diagnostic method in the
radiology department [12]. Moreover, to assist coronavirus-infected patients, dif-
ferent wearable monitoring devices such as medical devices respiration rate, heart
rate, temperature, and oxygen saturation, and respiratory assistive technologies
(ventilators, CPAP machines, and oxygen treatment) are routinely employed [13].
A patient can be fully monitored by a doctor using wearable technology during the
quarantine or self-treatment time. Wearable technology is the least expensive and
most often used option for social distancing. Any device issued by the company
could legally be tracked within the premises of the workspace, which could be
integrated with a site-wide notification system of potential health risks or any alerts
that adhere to the social distancing guidelines. A smart system could also be used in
conjunction with the wearable device to enable real-time and analytical solutions to
social distancing guidelines. A similar approach has already been implemented in
healthcare for the elderly, preventing potential hazards within the residences, such
as stairs, railings, and power cords. This technological answer has helped the
elderly in their daily lives; such devices monitor the person’s activity if they are
stationary or moving. If the surrounding area is mapped, then these activities could
be discretely identified as sitting or running. These aspects of wearable gear in
healthcare might serve as a model for developing social distancing technologies,
which enforce workplace standards and comprehend employee behavior, resulting
in the best outcome for everyone. However, the question of privacy still lingers
when presenting a perfectly manageable engineered solution, which is a bigger
concern when monitoring larger crowds in a bigger facility compared to a single
resident [14]. The concept is applied in research by integrating low-cost and ever
higher quality based metadata for patient monitoring. Future patient monitoring and
health information delivery systems will be dramatically impacted by advances in
communications combined with IoT and embedded system technologies [15]. This
can provide services through a smartphone theory of growth for chronic wound
image processing at the patient’s side, as well as providing better integration
between doctor and patient remotely, end-to-end daily schedule-based diagnostic,
and sustaining patient history [16].
7.4.2 Screening system

Most hospital screening methods before the COVID-19 circumstance allow for patients
to inadvertently come into contact with one another. Pressure measurement and pre-
liminary history recording may require being in close proximity to the ill individual for
less than 1 m, a distance that has the potential to be infected after the occurrence.
Because the number of medical workers is limited and cannot provide comprehensive
services, the world has focused on limiting exposure as a result of the pandemic. Various
sensor technologies have been used to assess patients before they begin treatments.
Many Thai hospitals include patient screening systems that are linked to the
central hospital database, such as a service queue booking system that allows
patients to get a queue before visiting the hospital, as detailed in the following
section. Patients can make an appointment to visit the hospital or simply stroll in.
Then they must pass via the hospital’s screening procedure. The hospital uses smart
video analytics to monitor temperature and can inform the nurse if the patient has a
high temperature. If the patient does not have a fever, the patient can proceed with
the screening procedure by using the following steps: a sensor or actuator for
weight and blood pressure measurement.
7.4.3 Queue systems

While the government is dealing with the COVID-19 crisis, several hospitals have
implemented screening systems using sensor devices. As the hospital has a limited
number of medical professionals, the hospital uses technology to assist with the
process, which queuing systems are increasingly being utilized in Thai hospitals.
The system connects with the hospital’s HIS system, which includes a registration
system. When arriving at the hospital, the queuing system software integrates with
a mobile application and a kiosk register, which uses an ID card in front of the
kiosk. Using the kiosk reduces social interaction and maintains the social distan-
cing procedure without the need to see a nurse. The system also includes a web-
based system accessible via a web browser. A queue system through a mobile app
is a method in which patients may use their phones to make appointments with
doctors in various departments, which can now be used to develop models for
predicting clinic selection. In terms of patient guidance, if the patient is still unable
to select which clinic to visit, the system may recommend a clinic and time for the
appointment. Furthermore, the system will communicate the patient’s information
to the examination room. The system will also transmit a LINE to alert patients
when it is nearing the end of the wait. As a result, the patient does not need to sit
Figure 7.4 Slip queue
and wait at the hospital for an extended period of time and may conduct errands
elsewhere first As a result, this method provides social distance control to minimize
the transmission of COVID-19. Figure 7.4 shows the slip queue that notified the
patient when the patient queue was nearly called.
7.4.4 Payment system

The payment system described earlier aids in cashless transactions, or if the organi-
zation wishes to decrease paper usage to save the environment. These computerized
payment methods can also assist in driving social distance. Because patients may
engage with the kiosk or electronic data capture (EDC) machine rather than the
cashier, the payment method has the potential to minimize the number of infections
during the COVID-19. Electronic payment methods are increasingly being employed
in some hospitals. The system will be linked to the hospital’s central database.
Furthermore, different forms of compensation, including social compensation and

government service, are available, as certain rights may be eligible for medical
assistance under the new payment system. Without the need to pay for medical costs
or medications, the use of electronic payment systems promotes efficiency and
reduces patient wait times during COVID-19. To provide additional electronic pay-
ment services, several Thai hospitals have expanded the number of payment kiosks
equipped with EDC devices to limit cash exposure and communication with hospital
cashiers. When the patient exits the examination room, they can proceed straight to
an electronic payment system to see if they are eligible for medical compensation
such as social security, government assistance, or others. In with the patient’s rights
and supports, the kiosk connects to the supporter database and clears the cost. If there
is an additional price for the case, the patient can pay it using an EDC machine or
cash. Using their ID card, the patient can pay by credit card, cash, or bank transfer.
They will receive the clearing bill if they are qualified for a reimbursement scheme.
The kiosk will produce a receipt to indicate that they have passed this process, and
the patient can go to the pharmacy section to obtain the prescription or other medical
necessities. Figure 7.5 shows the kiosk machine with an EDC reader.
Camera
Monitor
EDC
Barcode reader
Slip printer
Figure 7.5 Kiosk of EDC

Figure 7.6 LINE notifier
Patients will be assigned a number once they have made a payment in the
financial system. The LINE notification system will remind the patient within a few
minutes to pick up the medication. See Figure 7.6. Moreover, the patient can also
use delivery to deliver medicines at home.
7.4.5 Social distancing notified people in public

Thailand adopted the MorChana program as a tool to assist medical personnel,
government agency operators, and, most importantly, individuals in caring for and
protecting themselves from the COVID-19 pandemic. See Figure 7.7.
By tracking infection information, the program collects, utilizes, and dis-
seminates just the information required to avoid or suppress threats to each
individual’s health, as well as the others around them and the general public. The
system assesses chances or risks of infection and monitors the situation of the
COVID-19 outbreak by providing the information to the user. The program also
Figure 7.7 MorChana application
includes a QR code scanner for checking in and out of places such as shopping
centers or supermarkets, which would update the user’s timeline. If there were an
infected individual in the area, the application would send the information to the
Ministry of Public Health Database and notify the citizens of the immediate area
to evacuate the area. The update notification relies on mobile functions such as
GPS and Bluetooth to relay updates to users quickly. The infectious rate of the
virus decreased significantly with the use of the ThaiChana application. In the
future, this application may monitor the COVID-19 vaccine and others at risk of
becoming infected. Those who contact the disease avoid crowded locations,
where patients would be restricted to a particular area of mobility such as the
hospital or at home. The New Normal has proved to be a time of new technolo-
gical advancement with the advent of COVID-19, making technology a part of
everyday life. Not only can the latest mobile apps make life easy, but they also
aim to keep people safe and healthy. This issue government introduces an
application worth downloading and installing to help people deal efficiently with
present and potential developments.
7.5 Social distancing technology in manufacturing

Remote work is virtually never a possibility for production line employees, which
is not ideal for the COVID-19 situation. Manufacturing employees are still
adjusting to their jobs, with a particular emphasis on social distance as a result of
working in industrial factories. To prevent viruses from spreading on the manu-
facturing line, only a few personnel must be stationed on duty. A research study
opted to monitor worker behavior and detect breaches of social distancing guide-
lines on CCTV feeds using computer vision, resulting in real-time voice notifica-
tions on the manufacturing line. It is critical to keep personnel safe while still
keeping the business running. The research that follows is an efficient and cost-
effective approach to using AI to create a safe environment in an industrial setting.
It demonstrates how to develop a strong social distance evaluation technique by
combining contemporary deep learning and traditional projective geometry meth-
odologies [17]. The four technologies listed later have changed manufacturing
workforce management while keeping everyone safe in the process.
7.5.1 Checking the distance using wearable device

Wearable devices, as discussed before, are one of the common approaches when
implementing social distancing guidelines. Due to its lower cost, lower main-
tenance, and efficiency in tracking, it is deemed the most appropriate solution
within workspaces. Before COVID-19, wearable devices are used in healthcare for
monitoring patients, including symptoms and patient movements. Since these fea-
tures are nothing new and are used to assess health risk, it is easy to deploy in any
facility. Nowadays, digital tools and technical solutions play an important part in
detecting distance between various components of society, such as vehicle firms
employing sensors to avoid collisions. The audio tether can also be used to notify
employees if their personal gadgets are in close proximity to one another. Another
option is to attach a sounding device to helmets equipped with distance-sensing
devices, and light may be emitted as well. Another attribute that could be used to
monitor health risks is temperature. Products such as TempTraq or Oura ring are
wearable devices to measure skin temperature, which are used within the Cleveland
Medical Center for medical professionals who are working on the frontline. The
temperature sensors nowadays operate on low power, small in size, and they are
reliable and incorporated with existing wearable devices. The smart features, the
devices would be much helpful to detect abnormal body temperature changes. In
this case, it would be much easier to detect and isolate individuals before an out-
break began [18]. However, some plants in Thailand adapt existed technologies to
make social distancing between the workers. During the COVID-19, most of the
plants remained open, according to one interviewer. Employees must record their
travel or reveal the timeframe of 14 days before visiting the factory as a precau-
tionary step against infection. When workers request access to the facility, a rea-
sonable practice would be to provide a 14-day timeframe, maybe using an
application designed particularly for the company and for this purpose. Each zone
in the facility should be distinguished by partitioning or zone management, which

may decrease the risk of worker infection and product contamination.
7.5.2 Distance monitoring using Wi-Fi

Since Wi-Fi operates using radio signals, it is possible to locate heavy usage based
on the radius of the Wi-Fi. Some companies use existing Wi-Fi communication
systems to check Wi-Fi distances between people. It can view the location of
devices such as tablets, laptops, or smartphones and shows where each user is
frequented. Zigbee, for example, utilizes signal propagation to detect the location
of the users [19]. This approach allows employers to monitor employee positions.
The number of workers congregating in some areas employee movement and the
elapsed time in a particular location when one or more limits have been violated.
The system will display a notification. The advantage is that employers can identify
points of interest in their factories where large numbers of people gather and pro-
vide them with vital information to manage the distribution of workers in the fac-
tory, also saving costs. Because it uses the existing Wi-Fi communication system
and no upfront investments or employees wear sensors, no personal information is
collected [20]. Therefore, the privacy of the user is preserved.
7.5.3 Distance monitoring using video analytics

Some areas can use video analysis or high-resolution cameras instead of wear-
able devices or Wi-Fi-connected apps; companies may choose to use video
analytics. Using AI to monitor video and analyze the distance between people,
the system can alert supervisors when social distancing limits are violated.
Research on a real-time social-distance and pedestrian density detection system
was presented. Using modern video surveillance cameras and a deep learning-
based computer vision model can evaluate the pedestrian density and social-
distance patterns. Models such as YOLO v3 and DeepSort are deployed to
separate people from the surroundings and then identify them by individuals
with their bounding boxes. These boxes are then used to calculate the distances
between each person in real time, then the results are evaluated [21]. The pro-
posed technique for measuring social distancing patterns is based on the
assumption that by combining a pretrained object detection model, precise
range approximation, and post-processing filtering, one may get pedestrian
densities and range between each pedestrian pair [22]. The benefit is that video
analytics allow people to use their existing office camera system. It does not
require a smartphone or wearable app, while it requires investment, inspection,
and maintenance. The system can now identify social distance and deliver real-
time warnings. It can also predict the distance to the safe zone using data inte-
gration and machine learning. Management may utilize this information to plan
task management in the company. However, because video analytics imple-
mentation is pricey, some Thai businesses are unable to pay, and management
refuses to proceed in this fashion because it is unnecessary.
7.5.4 Social distancing by replacing some work with a robot

According to research, labor is one of the most important elements in manu-
facturing costs. If the firm wishes to expand its product line, it must hire more
workers throughout the manufacturing process. Investment in purchasing robots
to replace human labor has grown, but not to the extent that it should have. There
may be a large investment, but when the virus COVID-19 arises, individuals face
a high risk that impacts productivity and cannot address the problem even if those
individuals are vaccinated; nevertheless, another new virus strain may emerge in
the future. Both the client and the manufacturer are interested in intelligent
automation. People, particularly customers, seek alternatives where human
interaction is as minimized as possible. Furthermore, there are increasing auto-
mation products and solutions, which people of all levels have to access. Some
examples include self-checkout at grocery stores or online business systems,
which are more used apparently. Robots are also being used in airport hospitals
using robots for temperature screeners and food delivery bots to reduce manual
labor and induce social distancing. However, big-scale operations are also slowly
being replaced by robots and automation systems. Considerable cooperation such
as Amazon or Walmart has already started using robots to operate most of their
warehouses for shipping products. Sadly, most of those jobs replaced with auto-
mation, especially after the recession, are fully depleted for good. Therefore, the
decision to use robots to replace workers in some industrial plants tends to rise
significantly [23]. Even though the rise of automation is prevalent, the reliability
is not fully appreciated due to many factors such as the limitation to the tasks that
the robots are assigned or extensive data availability and proper usage. Indeed,
implementing robotics and automation has many benefits to many agencies dur-
ing COVID-19, but entirely relying on the nature of automation might not be a
worthwhile investment.
7.6 Social-distancing technologies for supporting

everyday life
The Johns Hopkins University and the Centers for Disease Control and
Prevention describe Social Distancing as a public health practice that isolates
infectious patients from healthy individuals in order to reduce the danger of
the virus spreading. Group activities should be canceled or done in enclosed
areas, and crowded venues should be avoided, in addition to being at least 2 m
apart from other individuals. The goal of social distance is to reduce COVID-
19 transmission. Suppose it is done correctly to reduce the risk of infection
among high-risk groups and healthcare professionals. In that case, the infec-
tion rate will be lower, resulting in fewer individuals being admitted to the
hospital.
Government policy in every country tries to introduce everybody to have social
distancing to decrease spread infections that are as follows.
7.6.1 Technologies support working at home

7.6.1.1 Remote working
The spreading and deadly coronavirus, also known as COVID-19, has profoundly
impacted the global economy. All countries are trying to find alternatives to deal with
this challenging situation. There is an urgent need to protect employees in offices. In
Thailand, both the public and private sectors have adopted Internet-based front-end
technology as an integral part of their services for employees and public users.
COVID-19 compelled every country to enter the New Normal phase, which may not
be optimal for countries where technology is scarce. As a result, digitizing all aspects
of the organization is difficult. Working from home still has network security con-
cerns while transitioning to the New Normal because developing countries’ network
structure and design are inadequate. As a result, network security problems have
become increasingly frequent since beginning work from home; most work from
home is not linked through virtual private network (VPN) and still uses the Internet.
As working from home becomes more critical and security features are introduced, a
VPN is built on a packet-based network and consists of some selected packet-based
network resources. VPN is the most simple and efficient tool for people to safeguard
their Internet data and conceal their identity online. When a device connects to a
secure VPN server, the user’s Internet data is routed through encrypted tunnels that
no one can see, including hackers, governments, and ISPs. However, VPN technol-
ogy and support systems are not without security dangers and risks. People and
business persons who are working from home use computer networks and virtual
environments that hackers or cybercriminals have abused. One of the biggest chal-
lenges facing any organization in adopting VPN technology is VPN gateway location
and access control. It is a policy challenge directly related to the nature of the net-
work security infrastructure that organizations deploy to secure knowledge, appli-
cations, and networks. The remote access technology and security checklist consist of
13 main security feature categories the organizations provide to their staff, admin-
istrator, and authorized users. Details of each category within the checklist are dis-
played in Tables 7.1 and 7.2.
7.6.2 Applications support work from home (WFH) service

7.6.2.1 Call center
Call center personnel are now regarded as the organization’s beating heart. Call
centers keep businesses running, with no set work hours and few holidays to
guarantee that customers are satisfied. However, suppose call center employees are
required to remain up late at night. In that case, there will be issues with the safety
of returning home, and the majority of them work in congested areas that are vul-
nerable to COVID-19 spread. The call center turnover rate is quite high; the firm
requires technology that allows employees to work from home in order to maintain
social distance. Voice over IP (VoIP), often known as Voice over Internet Protocol,
is a means of communicating over the Internet using voice. It is a novel method of
making calls via the Internet network, with the first evident advantages. Because
utilizing the Internet to make a phone call eliminates the need to travel via a
Table 7.1 Remote access security feature categories
Number Category name Description

1 General organization remote access ● Staff remote access
instruction available to their remote ● Administrator remote access
access users
● Authorized users remote access
2 Organization provides remote access ● Staff remote access
hardware to its authorized users ● Administrator remote access
● Authorized users remote access
3 The organization has a different ● Administrator
group’s level of remote access users ● Staff
● Other authorized users
4 Remote access technology ● Customer managed solution
● Virtual private network (VPN) and
encryption methods
● Cloud-based technology
5 Central remote access architecture, ● Remote access gateway architecture
mechanism, and integration system ● Gateway DNS name
● Remote access mechanism
● Remote access system integration
6 End device software and system re- ● Operating System (OS) support
quirements ● User remote access software
application
7 End device security control ● Operating System (OS) security
● Use well-known strong personal fire-
wall/antivirus/malware
● Web browser security
(clientless mode)
● Vendor or built-in remote access cli-
ent software (IPSec/SSL)
8 Central site authentication technology ● Employed encryption and digital
certificate technologies
9 User site authentication technology ● Logon to remote access server
requirement
● Using two-factor (2FA) authentica-
tion as second security logon layer
● Logon failure limitation
● Logon user input type
● Password restriction/requirement
● Single sign on (SSO)
10 Central site network access and system ● Pre-logon
security ● Host Information Profile (HIP) control
● Internal firewall in-place
● filtering
● Block malicious domains (DNS
filtering)
Table 7.2 Remote access security feature categories (Cont.)
Number Category name Description

11 Central site application ● Client image application (i.e., Citrix, VM)
security ● Automatic timeout feature for inactivity Web brow-
ser/Client Software/Idle timeout
● Session timeout
● Multiple concurrent sessions
● Bandwidth rate limiting to priority remote
access users
● Password policy management
● Logging information and alert
● Vulnerability protection
● File blocking
● Deep content inspection
12 IT assistance, monitor- ● Hotline/helpdesk service availability for
ing, and support personal help
13 Remote access IT policy ● Use of remote access and Internet usage policy
and procedure ● Use of information security policy (remote access
related) to the university’s staff and lecturers
● Use of information security policy—technical to the
university’s IT staff
● Regular update or review the remote access
related policy
● Advise general remote access and
● Internet usages, including security awareness to new
remote access user
● Frequent advice to staff, lecturers, and students for
IT security information including remote access,
Internet related
telephone network, it can help reduce the cost of calling, whether it is an interna-
tional or domestic call. IP network communication requires the employment of a
router (Router) with a specific purpose to ensure the quality of the IP channel in
order for the information to be sent or returned correctly. Furthermore, there may
be privileges before other IP packets (quality of service: QoS) to provide high-
quality services, which will be seen in the development of current network equip-
ment switches, which have included Port QoS. Employees at businesses utilized
VoIP phones; when working from home, they may continue to use the phone as if
they were at work. A program named SIP Client must be installed on the mobile
system. To communicate with the device, the user may also utilize the SIP Client
program if they wish to use it from their computer.
7.6.2.2 E-Documents
During the COVID-19 epidemic, medium-sized and big companies prioritized IT
systems, working on papers online. Electronic Document Management System
(EDMS) improves the organization’s working system by making it more con-

venient, faster, and efficient. It reduces the time spent searching for documents,
which reduces the number of staff in operations and the cost of purchasing docu-
ment management resources. It also reduces the amount of paper used. The benefits
of adopting an EDMS include the ability to effectively add, remove, modify, and
search documents saved in the database. It is simple and quick to use the software
installed on the computer or operate through the browser at the organization. As a
result, there is no issue with the original document being destroyed or damaged as a
result of poor storage, poor quality, or loss without anybody being able to discover
it. System security regulates system usage by granting rights to different groups of
users, such as admin groups, data collector groups, and general user groups. EDMS
also backs up the work, and the database may be utilized in the event of a system
crash or damage as well.
7.6.2.3 Cloud service

The term “public cloud” refers to a cloud service that is available to anybody.
Dropbox, Amazon, OneDrive, and Google Drive are all examples of completely
public cloud services. Everyone who uses the public cloud is renting a shared server
space. As a result, the public cloud may be thought of as a shared environment,
similar to information stored in the main office. Private clouds vary from public
clouds in that all virtual machines and cloud infrastructures are exclusively avail-
able to the customer. Everyone has access to anything on the Internet, but the server
hosting only corrects the customer data. Because the server holding your data is not
shared among numerous customers, some people may want to utilize a private
cloud for added security, whilst others may require a personal cloud for increased
productivity. The private implementation may need a significant expenditure, such
as a data center, server, and IT administrator. In addition, unlike the public cloud,
the private cloud provides clients control over how their servers are managed, kept
secure, or entirely backed up. Most people use public clouds; however, private
clouds are ideal for individuals that manage large amounts of data on a large scale.
Using cloud services is a technology that may embrace social distance by allowing
employees to work in multiple locations without limits while also decreasing
congestion among employees who must work in the same location. As a result, the
proliferation of COVID-19 is reduced.
7.6.2.4 Management information system

A management information system (MIS) is a system that gives executives the
information they need to work effectively with both internal and external data. The
data is about the firm, both in the past and in the present, as well as what is
expected in the future. Furthermore, the MIS system must provide data at key times
so that administration may make sound decisions about guaranteeing the security
and functioning of the company. Although the highest MIS system will benefit
middle management the most, the MIS system will be a system that can assist
information executives at all three levels: top-level executives, middle manage-
ment, and senior management. The MIS system will create a report that is a
summary of data gathered from all corporate databases. The objective of the report
is to provide executives with an evaluation of current developments, an overview of
the company, and drive management and audit operations. On the other hand, the
report’s breadth is defined by the nature of the data and its intended purpose.
Reports may be issued on a regular basis (e.g., an income statement or a balance
sheet), on-demand, or in response to specific situations or occurrences. An MIS can
be designed to allow the user to access anywhere by accessing through a web
browser and the MIS client system installed on the computer. To safeguard access
to information, the user may need to utilize a VPN. MIS is a technology that allows
people in many departments to work from home as if they were in an office.
7.6.3 Conferencing application

Due to the extreme COVID-19 virus outbreak, several firms are requiring their
workers to work from home. Although online meetings are popular, they are not as
productive as face-to-face meetings. Furthermore, many organizations’ policies are
to begin work from home procedures, participate in conferences, and hold meet-
ings, whether in or out of the office. Video conferencing has become the standard,
and with it, online meetings and seminars will become part of the New Normal
workplace culture. This does not just apply to working in a corporation but also to
study at a university or a high school. Many people, even those with impairments
who need to be with their families, may benefit from this technology. This is not
just an in-company only, but studying in the university or school as well. This tool
may benefit many people, such as people with disabilities and who need to be with
their families. Today, practical meeting tools for social distance are widely used, as
described in the following subsections.
7.6.3.1 LINE
Thailand already had a high number of LINE users prior to the COVID-19.
However, it is not yet obvious for academic purposes. After the lockdown, LINE
has played a larger part in education, and teachers must develop ways to commu-
nicate with pupils about homework and assignments. All the school directors,
instructors, and educational staff interviewed stated that by using the group LINE
to establish a group with students, LINE returned to play a role in teaching and
learning. A new LINE feature is the ability to use group calls that appear like
teleconferencing. LINE groups make it easy to arrange online courses. Online
tutoring through LINE’s group sessions, on the other hand, is exclusively offered to
primary school pupils. Teachers will seldom utilize LINE with kindergarten chil-
dren. Teachers may only meet with pupils for 10–20 min so that kids may go to
activities or worksheets, and parents can send worksheets to the teachers via LINE.
In addition, LINE also plays a role in many other meetings because it can be used
for unlimited meetings. Furthermore, a small group can meet together without
many of the advanced features, allowing for communication individually. Parents
could conveniently send files and images for meetings. Both large and small
entrepreneurs use LINE as their main communication service since the lockdown.
7.6.3.2 Zoom
In Thailand, many universities consider using Zoom applications. The user who has
created the room is considered an administrator or host of the room. When a person
enters a Zoom conference room, they are considered an attendance hosted by another
user. The meeting organizer or substitute host (if the original host is unable to attend)
will have host controls, such as the capability to silence audio, utilize video, share
your screen, and more. The attendance microphone can be muted and unmuted by the
host. The host can also control the room’s sound by clicking on the arrow next to the
mute button, and the host will get further audio settings options. The host, for
example, can change the attendance microphone, switch off the computer’s audio, or
access the audio settings. Attendees can switch on or off the camera by pressing the
Start/Stop Video button. By pressing the arrow adjacent to the start/stop video button,
attendees can swap showing cameras, adjust Zoom video options, or change a virtual
background. The attendance can see who is presently in the meeting by clicking on
participants and raising their hand or changing the name on the participant’s list.
Moreover, the attendance can share their screen in the meeting if the host allows it.
The attendance screen share will be able to be stopped by the host. Other features of
Zoom include a breakout room assignment. Zoom allows the host to assign partici-
pants into groups which encourages a smaller and active learning environment. The
host or the instructor could quickly assess each student of their understanding, which
is proven to be a convenient online alternative learning experience for college taught
subjects that require active participation and higher level understanding, such as
medicinal chemistry [24].
7.6.3.3 Google Meet
During the COVID-19 epidemic, the Google Meet application provides secure
video conferencing for teams and companies while remaining simple for students to
utilize. Organizations may use enterprise-grade video conferencing based on
Google’s secure global infrastructure to keep employees securely connected and on
the go. It is backed up with safe design architecture.
Many Thai institutions have Google Education licenses, which are part of the
Google Meet platform. Google Meet uses self-security and a well-known world-
wide network to secure data and privacy to the greatest extent feasible. During
transmission, Google Meet uses an encrypted video session, and various anti-abuse
measures are activated by default. Google Meet is offered at no cost to educational
institutions. However, in order to host a virtual meeting for up to 250 participants,
the institution must register as an enterprise. Google Enterprise also includes fea-
tures such as internal live broadcasting and the ability to save call recordings to
Google Drive and the ability for room creators to invite up to 100 people to each
Google Meet session without requiring them to install the Google Meet app on
individual devices. Users can join the conference using only a web browser;
moreover, room creators in Google Meet Enterprise can add group members for up
to 24 h of video chatting. Meeting room organizers may now view all of the day’s
scheduled meetings in one place and join with a single tap from the dedicated
Gmail mobile app or Meet app. Google Workspace also assigns a phone number to
each meeting, ensuring that all attendees have a positive experience. Meetings may
be joined immediately via calendar events, email invitations, or Gmail, which is
also linked to Google’s infrastructure. Those in attendance can join the meeting
from any computer, phone, or conference room. All event details will be consistent.
With live captioning, low-light mode, and noise suppression are employed, meet-
ings become more productive.
7.6.3.4 Webex
Customers may need to hold an online conference when they begin to work from
home. The school should invest more in online instruction and has begun to employ
video conferencing software for distant sessions. During this time, one of the most
popular programs is “Webex,” which can hold continuous meetings for 50 min and
handle 100 devices. Webex may be accessible via a variety of channels, including
installing an app on a Tablet or Smartphone for both iOS and Android or using a
web browser on a PC, Windows, or MAC OS, and there is software available to
download into the device for convenience.
7.7 Social distancing and smart city

7.7.1 AI and big data
Various sensors and IoT devices are instruments that might be quite useful during
the COVID-19 epidemic. Whether via shortwave radio networks such as Wi-Fi,
NB-IoT, or cellular, these devices serve to notify the distance between individuals;
if these devices join with the Internet system across the city, data from these
devices may be gathered for processing. Furthermore, the data collected may be
utilized with machine learning to develop a model to anticipate various clusters and
infections and utilize GPS to notify individuals in close vicinity by putting them on
a map or even creating a model for vaccine planning. The information might be
used to create a quarantine plan for affected people or even to allocate hospital beds
to deal with future epidemics. Thailand, for example, has a MorChana application
that can monitor the course of infection from one person to another and is used for
vaccine registration, which may be assigned to multiple medical facilities. Big data
refers to information gathered through data gathering across the country and is used
to plan vaccine allocation. The data can also be used to forecast the number of
epidemic infections in the future by developing a machine learning model, such as
linear regression or artificial neural network, with the input layer consisting of the
number of infected people at the time, the number of classes, the infection rate, the
vaccination rate, the death rate, the number of hospital visits, and the number of
people who have been screened.
7.7.2 Implementation and usability

The COVID-19 problem in Thailand prompted application developers to construct
an application (MorChana) for the government in a short period, despite the pub-
lic’s lack of knowledge of the usage. User-friendliness and UX/UI compatibility are
critical for the major aspects of the application, which allow users to encounter
minimal issues in order to receive correct and timely information. Furthermore, the
adoption of this technology across the country causes frequent network failures and
the delivery of erroneous SMS messages, resulting in the message not being ver-
ified. There was an issue with data handling, and the service had to be terminated.
These applications rely on Google Maps for network services in order to use the
maps. Furthermore, alerts from LINE apps and Facebook improve efficiency.
7.7.3 Privacy and security

When a huge volume of data is gathered, it is possible that it will contain illegal
personal information. It is an issue in terms of violating personal rights, regardless
of its stated goal of giving useful information for tracking the spread of COVID-19.
As a result, privacy checks and the partial deactivation of personally identifiable
information are necessary. Another consideration is the exchange of information
across organizations. The link between government agencies such as hospitals,
schools, subdistrict administrative organizations, and volunteers, the information
transmitted may contain personal information that will impact that individual in the
future.
7.7.4 Policy and legislation

Thailand passed the Personal Data Protection Act in 2019 to protect the personal
information stored by government agencies. Citizens will be allowed to defend their
privacy under the new law. The law, however, has been delayed for another year and
will go into effect in 2022. Opening personal data regarding travel routes, work, and
places is a breach of personal data, even if the individual did not intend to expose the
information. As a result, after COVID-19, the government must account for data
utilized for future epidemic prevention. During the COVID-19 period, people
experience a variety of emotional reactions, including fear and anxiety. Because of
the cognitive effect, COVID-19 has an impact on purchase decisions during lock-
downs. One study examined the content of online newspapers and concluded that it is
an important emotional component in individual purchase decisions. Incorrect belief
rumors/false news, various types of viral social media news, and survival psychology
are all key cognitive variables [25]. Furthermore, resolving the COVID-19 infection
must not infringe on individual rights and must be supported by legislation in each
nation. Major firms agreed that partial information must be disclosed in the form of
an aggregate that does not identify a person. This information must be used to
observe the distribution of COVID-19 clusters or other information that is used
ethically.
7.8 Conclusion and future works

Amidst the unexpected crisis that arises from COVID-19, humanity looks for
technology to solve problems, especially in their daily lives. Education is the first
aspect where technology could be used to mitigate the health risks of crowded
classrooms and auditoriums, but it also presents alternate solutions for those who
seek further knowledge and fill educational gaps. LMSs are used widely by edu-
cational institutions to ease the access of course materials and to provide a simple
tool for instructors to use; examples mentioned include Google Classroom and
Moodle. MOOC is another alternative to regular classroom settings, allowing stu-
dents to be able to access university-level course content online for free, which
promotes self-learning during the closure of schools and university campuses due
to COVID-19. Along with the education sector, healthcare is in need of technolo-
gical solutions to manage social-distancing guidelines in facilities and the man-
agement of patients. Wearable technology becomes a commodity for both patients
and healthcare workers, which allows the administration to track movements and
location and notify if a health policy has been a breach or if there is a potential
health risk. In large healthcare facilities, a system is put in place to guide patients
through the medical process easily. This includes a screening system to scan
patients upon arriving at the facility, a queue system to manage the order of patients
receiving medical care, and a payment system to ensure a safe and efficient
checkout. However, protecting the public interest in health and safety from
COVID-19 extends beyond the hospital walls. Some applications manage the
general population’s well-being by notifying them of potential health risks and
organizing other health services in one system, such as vaccination registration.
Another sector that employs the use of technology is manufacturing. Some tech-
nological solutions include using big data and AI such as video analytic and Wi-Fi
monitoring to mitigate infection rates within manufacturing facilities. Some man-
ufacturing firms also require employees to have a wearable device to monitor their
conditions and their location for the best of the employee’s interest and safety in the
workspace. While health safety is important for the human workforce, intelligence
systems are being used in all systems where human labor is redundant. These
systems not only reduce the risk of mass infection within factories but also save
cost and time. Nevertheless, many firms allow employees to work from home,
which will help reduce employee interaction. To ensure protection from a data
breach or other technological threats while working from home, VPN access and
entitlements are given to each employee. Access to sensitive information is very
limited. Different applications that would support remote work are also mentioned.
These include VoIP, e-documents apps, cloud service, MIS, and conferencing apps,
such as Zoom, Google Meet, and Webex. VoIP is very beneficial for call center
applications, where existing Internet service and all voice calls are processed
through the same channel. E-documents applications enable online collaborative
documents to work to be done remotely, which could be a part of cloud services
where file-sharing is made easier through the cloud. Meetings and conferences
could also be held online due to in-person limitations by apps such as Webex,
Zoom, and Google Meet. Achieving a fully hi-tech “New Normal” society requires
a smart city framework. Agencies on many levels tend to lean in on AI and big data
solutions. Still, intelligence systems tend to have limitations, and understanding the
usage would be key to solve necessary tasks. The understanding of technology
should also be given to the general public if technology is being implemented for
their purpose. Other factors that should be considered are the privacy and security
of the users, which are required by laws and policies.
References
[1] Muhammad L., Algehyne E.A., Usman S.S., et al. Supervised Machine
Learning Models for Prediction of COVID-19 Infection Using Epidemiology
Dataset. SN Computer Science. 2021; 2(1): 1–13.
[2] King N. Template Analysis. Sage Publications Ltd; 1998.
[3] Brooks J., McCluskey S., Turley E., et al. The Utility of Template Analysis
in Qualitative Psychology Research. Qualitative Research in Psychology.
2015; 12(2): 202–222.
[4] Gunawan G., Suranti N.M.Y. and Fathoroni F. Variations of Models and
Learning Platforms for Prospective Teachers During the COVID-19
Pandemic Period. Indonesian Journal of Teacher Education. 2020; 1(2):
61–70. Available from: https://journal.publication-center.com/index.php/
ijte/article/view/95.
[5] Raza S.A., Qazi W., Khan K.A., et al. Social Isolation and Acceptance of the
Learning Management System (LMS) in the Time of COVID-19 Pandemic:
An Expansion of the UTAUT Model. Journal of Educational Computing
Research. 2020; 59(2): 183–208. Available from: https://doi.org/10.1177/
0735633120960421.
[6] Octaberlina L.R. and Muslimin A.I. EFL Students Perspective towards
Online Learning Barriers and Alternatives Using Moodle/Google Classroom
during COVID-19 Pandemic. International Journal of Higher Education.
2020; 9(6): 1–9. Available from: http://www.sciedu.ca/journal/index.php/
ijhe/article/view/18485.
[7] Hermanto F. and Putri N.A. Learning Ability and Supporting Capacity of
Online Learning In Students during Covid-19 Pandemic. 3rd International
Conference on Environmental Geography and Geography Education IOP
Conf. Series: Earth and Environmental Science. 2021; 747(1): 012080.
Available from: https://doi.org/10.1088/1755-1315/747/1/012080.
[8] John C., Staubitz T. and Meinel C. Took a MOOC. Got a Certificate. What
Now?. In: 2019 IEEE Frontiers in Education Conference (FIE); 2019. p. 1–
9. ISSN: 2377-634X.
[9] Razai M.S., Kankam H.K.N., Hourston G.J.M., et al. Lessons Learned From
Producing Guidance Articles and Rapid Massive Open Online Courses
During the Covid-19 Pandemic in Primary Care. Journal of Primary Care &
Community Health. 2020; 11: 2150132720963624. Available from: https://
doi.org/10.1177/2150132720963624.
[10] Sutrisno S. Increased Learning Activities and Outcomes Through Online
Learning with Google Classroom in The Covid-19 Pandemic Period. Ideguru:
Jurnal Karya Ilmiah Guru. 2020; 5(1): 95–106. Available from: https://jurnal-
dikpora.jogjaprov.go.id/index.php/jurnalideguru/article/view/151.
[11] Filippidi A., Tselios N. and Komis V. Impact of Moodle Usage Practices on
Students’ Performance in the Context of a Blended Learning Environment.
Proceedings of Social Applications for Life Long Learning; p. 2–7.
[12] Chakraborty C. and Abougreen A.N. Intelligent Internet of Things and
Advanced Machine Learning Techniques for COVID-19. EAI Endorsed
Transactions on Pervasive Health and Technology. 2021; 7(26).
[13] Islam M.M., Rahaman A. and Islam M.R. Development of Smart Healthcare
Monitoring System in IoT Environment. SN Computer Science. 2020; 1(3):
185. Available from: https://doi.org/10.1007/s42979-020-00195-y.
[14] Alarifi A. and Alwadain A. Killer Heuristic Optimized Convolution Neural
Network-Based Fall Detection with Wearable IoT Sensor Devices.
Measurement. 2021; 167: 108258.
[15] Chakraborty C., Gupta B. and Ghosh S.K. A Review on Telemedicine-Based
WBAN Framework for Patient Monitoring. Telemedicine and e-Health.
2013; 19(8): 619–626. PMID: 23841489. Available from: https://doi.org/10.
1089/tmj.2012.0215.
[16] Chakraborty C., Gupta B. and Ghosh S.K. Mobile Metadata Assisted
Community Database of Chronic Wound Images. Wound Medicine. 2014; 6:
34–42. Available from: https://www.sciencedirect.com/science/article/pii/
S2213909514000445.
[17] Khandelwal P., Khandelwal A., Agarwal S., et al. Using Computer Vision to
enhance Safety of Workforce in Manufacturing in a Post COVID World;
arXiv:2005.05287. https://arxiv.org/abs/2005.05287.
[18] Seshadri D.R., Davies E.V., Harlow E.R., et al. Wearable Sensors for
COVID-19: A Call to Action to Harness Our Digital Infrastructure for
Remote Patient Monitoring and Virtual Assessments. Front Digit Health.
2020; 2: 8.
[19] Nguyen C.T., Saputra Y.M., Van Huynh N., et al. Enabling and Emerging
Technologies for Social Distancing: A Comprehensive Survey and Open
Problems. IEEE Access. 2020; 8: 153479–153507. Comment: 43 pages, 21
figures, journal.
[20] Venkatnarayan R.H. and Shahzad M. Demo: Measuring Distance Traveled
by an Object Using WiFi-CSI and IMU Fusion. In: 2019 IEEE 27th
International Conference on Network Protocols (ICNP); 2019. p. 1–2.
[21] Punn N.S., Sonbhadra S.K., Agarwal S., et al. Monitoring COVID-19 Social
Distancing with Person Detection and Tracking via Fine-Tuned YOLO v3
and Deepsort Techniques. arXiv:200501385 [cs]. 2021.
[22] Zuo F., Gao J., Kurkcu A., et al. Reference-Free Video-to-Real Distance
Approximation-Based Urban Social Distancing Analytics amid COVID-19
Pandemic. Journal of Transport & Health. 2021; 21: 101032. Available from:
https://www.sciencedirect.com/science/article/pii/S2214140521000268.
[23] Coombs C. Will COVID-19 Be the Tipping Point for the Intelligent
Automation of Work? A Review of the Debate and Implications for
Research. International Journal of Information Management. 2020; 55:
102182.
[24] Singhal M.K. Facilitating Virtual Medicinal Chemistry Active Learning

Assignments Using Advanced Zoom Features during COVID-19 Campus
Closure. Journal of Chemical Education. 2020; 97(9): 2711–2714.
[25] Roy S. and Chakraborty C. Panic Buying Situation During COVID-19
Global Pandemic. Journal of Information Technology Management. 2021;
13(2): 231–244. Available from: https://jitm.ut.ac.ir/article_80625.html.
Chapter 8
Social health protection in touristic destinations
during COVID-19
Zaklina Spalevic1, Aleksandra Stojnev Ilic2 and Milos Ilic3
Abstract
COVID-19 pandemic has caused huge economic losses in almost all industries.
Thus, the economic crisis caused by this pandemic has modeled huge challenges
for leisure industry. Travel bans, closed borders, mandatory quarantine after border
crossing, reduced contacts are just some of the problems that citizens around the
world have encountered. Observed from the point of view of tourism, a special
problem was curfew and quarantine closure of cities and areas. All these sig-
nificantly disrupted global tourism, which have been in full upswing in the years
before pandemic. The research described in this chapter aims to give a description
of the proposed software solution intended for better control and records of infected
persons. The straightforward use of the proposed software solution aims to reduce
the potential for the virus to spread among tourists. In order to reduce the potential
for the spread of the virus, the proposed software solution needs to enable a better
control of the movement of infected persons. Proper control of the movement of
infected persons reduces the pressure on the tourism sector and opens the possibi-
lity of reoperation of tourist facilities for persons who are not infected. The pro-
posed system is based on mutual cooperation of several actors. Thus, at the local
and state level, the healthcare system can be singled out. Actors acting only at the
local level are tourism organizations and local governments, while the actor acting
only at the state level is border control. When it comes to the process of collecting
personal data of patients, the bearer of this task within the system is the local health
organization, which is certainly the first in contact with infected patients. In this
way, the role of the healthcare organization is the proper registration of infected
patients, and their entry into the proposed system. This part of the system also
provides data on vaccinated and revaccinated patients. In order not to further bur-
den the healthcare system, proposed software solution will use existing data and
1
Faculty of Tourism and Hospitality Management, Singidunum University, Belgrade, Serbia
2
Department of Computer Science and Informatics, Faculty of Electronic Engineering Niš, University of
Niš, Niš, Serbia
3
Department of Computer Science and Informatics, Faculty of Technical Sciences, University of
Pristina, Kosovska Mitrovica, Serbia
will not require additional tests. Based on provided data from all actors, the inte-
grated management system will create a database with vaccinated or tested positive
persons. This database will be obtained by merging several databases into one so-
called centralized register. However, each of the actors in the system has the right
to use only that part of the data that is essential for their work. In this way, the
system is able to protect personal patient data from unauthorized access. The sys-
tem will allow one to check vaccination or home isolation status of an individual
and, based on the results, approve or deny reservations. The edge/cloud architecture
at the same time will provide data availability, and data access control.
Keywords: e-Health; COVID-19; Cloud; GDPR; Software application
8.1 Introduction
When this software solution is proposed (May 2021), 168 million people were
infected by COVID-19 virus. In the same time this virus caused the deaths of
around 3 million people worldwide. From the start of the 2020 year, more precisely
after the onset of the COVID-19 virus, a mass pandemic people’s daily activities
were disordered with different constrains and rules. People’s daily activities like
interacting with each other, traveling, having fun have taken on a whole new form
or have been banned. COVID-19 also affects different fields of world economy.
From the aspect of human health, the COVID-19 virus has endangered both the
physical and mental health of the world’s population. In addition to physical and
mental health, the fight against the pandemic also affected personal data privacy.
Personal data were found in one chain cycle, where on the one hand it is necessary
to preserve those, while, on the other hand, it is necessary to enable their processing
in order to reduce pandemic [1].
One of the ways to fight the pandemic proposed by the World Health
Organization is to close the borders of many countries around the world. Their
borders remained closed for a long period of time for both visitors and tourists. This
move has caused great economic losses, especially for those countries that base
most of their budgets on tourism. Furthermore, the pandemic affected the tourism
sections of these countries significantly, making it harder for economy. For the first
time ever all global tourist destinations have introduced some form of restriction
[2]. Restrictions in terms of movement, gatherings, travel did not bypass domestic
tourism either. Revenues from domestic tourism have decreased significantly. The
easing of measures enabled the reopening of tourist destinations. Of course, the
emphasis on reopening was placed on respecting the prescribed measures and
respecting the imposed restrictions. The overall situation with the pandemic has led
to the fact that there are a large number of individuals who, despite the easing of
measures and the opening of borders, do not dare to go on international trips. There
is also a group of people who cannot afford international travel given that they have
lost their jobs or their incomes have been significantly reduced. Prior to the
Social health protection in touristic destinations during COVID-19 211
pandemic, tourism was considered the sector with the highest economic growth. In
support of this are the data that in 2018 alone, 1,407 million international depar-
tures were made. For the sake of comparison, some of the most popular countries in
terms of the number of tourists such as France, China, Spain, during 2020 faced a
large number of infected patients and the closure of borders, which led to a huge
drop in visits compared to 2019. The extent to which the pandemic caused eco-
nomic losses to the tourism sector is also shown by the fact that a loss of $297
billion was recorded in the United States alone, as a result of the suspension of
tourism in the first quarter of the year [3]. Furthermore, for a very large number of
countries, tourism represents the main income for their citizens, and in the same
time basic government revenue, based on the fact that tourism is one of the
industries that employ a significant number of workers. Without such a key type of
earnings, many countries experienced significant losses in gross domestic product.*
One of the partial solutions for economic and psychological crisis is to gather
necessary information about tourists and use them to allow or deny different
activities. By providing a solution for simplification and organization of patient
data, categorization of individuals tested positive or negative, without compro-
mising privacy rights. All these will be achieved within the proposed software
solution by introducing different privacy policies. This will provide greater travel
opportunities for individuals tested negative for COVID-19 or vaccinated ones.
Viewed from the point of view of patients whose results have shown to be positive
for the virus, the proposed software solution enables a better control of the move-
ment of such patients, as well as a better control of compliance with measures to
prevent the spread of the virus.
The proposed software solution described in this chapter is based on the use of
data obtained from different sources. The data obtained from the health system
database represent the backbone of the entire system. In practice, these data are
crucial in separating people positive and negative for the COVID-19 virus. The
data obtained from the health system within the proposed software solution are
combined with the points obtained by the local administration, tourist organiza-
tions, as well as border control. In this way, the possibility of monitoring the
movements of infected persons is realized. The contribution of the proposed system
is primarily reflected in providing better data protection for persons registered as
COVID-19 positive. In addition to data protection, the proposed software solution
should reduce the spread of the virus by ensuring a better monitoring of infected
persons and preventing their reckless behavior. Finally, for all those who are not
infected, as well as for tourism in general, the contribution of the software solution
is to ensure the reopening of tourist destinations and the free movement of all
persons who are negative for the COVID-19 virus or vaccinated against it.
The chapter is organized as follows: the second section of the chapter provides
an overview of the literature as well as relevant papers related to the research
presented in this chapter. The third section describes the architecture of the pro-
posed software solution intended for use in order to reduce the spread of viruses
*
OECD Better Policies for Better Lives, Retrieved 15.10.2020, from http://www.oecd.org/.
and increase opportunities for tourism recovery. An overview of legal norms, laws,
and bylaws related to the protection of personal data during the pandemic of the
COVID-19 virus is given in the fourth section. The fifth section of the chapter
summarizes key benefits of proposed system, key conclusions, and plans for future
research.
8.2 Related work
The traditional approach to monitoring the health status of patients, as well as their
treatment, has changed a lot during the last 2 years. A large number of medical
institutions have switched from the traditional to the remote approach to treating
patients, in all those cases where this was possible. A significant number of studies
conducted on this topic have contributed to this treatment system. One of the
researches on this topic described in [4] presents system for home healthcare.
Practically, this is a system with remote access whose task is to provide assistance
in the healthcare home. Within this chapter, the authors proposed a software system
whose task from the patient’s point of view is to obtain a prescription for medi-
cations without going to the doctor. System enables monitoring of the patient’s
health status in order to determine the right therapy. The software solution is based
on the Android mobile application. This application connects to a web-based
application. In this way, mutual communication between the doctor and the patient
is possible. The collection of data relevant to the determination of therapy is done
automatically using sensors. These sensors record the psychophysical parameters of
the patient. In order to provide accurately read data of psychophysical parameters,
the authors performed the installation of a HAC monitoring framework. With the
implementation of this framework, the performance of the system itself has been
improved. The proposed software solution aims to monitor patients remotely,
without any contact with them. The proposed mobile platform enables the sending
of daily values of parameters, as well as communication between the doctor and the
patient, during the period of time that the patient spends in isolation.
The use of telemedicine to improve health services in rural and highly occu-
pied areas of India was one of the topics of research presented in [5]. Within this
chapter, the authors emphasized the possibilities of using new technological solu-
tions in order to improve the healthcare system of India. The authors also empha-
sized that the use of technological solutions is especially visible during epidemics
and pandemics. Epidemics and pandemics pose a kind of challenge for both the
healthcare system and healthcare workers. The conclusion of the author is that the
need for new information and communication solutions leads to quality changes in
the management of the healthcare system. The research highlighted potential
challenges, as well as the opportunities offered by the use of telemedicine in order
to improve the healthcare system during a pandemic.
In the study conducted by the group of authors, a system aimed to detect
COVID-19 virus based on thermal image processing is shown. Reduction of
interference among people is provided by the use of a smart helmet [6]. The
technologies on which the proposed solution is based included a thermal camera

built into the smart slam and IoT technology. The task of IoT technology is reflected
in monitoring the scanning process itself, as well as collecting data in real time.
Also, the use of the mentioned technologies enabled another functionality that is
reflected in the impregnation of faces. Practically, by impregnating the face, it is
possible to obtain personal information of the person who is being scanned, and
whose temperature is being measured. Such a system can be especially important
when it comes to scanning the temperature of passengers at border crossings, as well
as at airports of known tourist destinations.
Author in [7] presented the implemented information and communication tech-
nology system for remote monitoring of intensive care unit. The operation of the
proposed system is based on the use of specially designed cameras and smartphones.
The use of closed-circuit television cameras is suggested. Such high-resolution
cameras are installed throughout intensive care. More precisely, a camera that
monitors the operation of mechanical ventilation is placed above each bed in the
intensive care unit intended for patients. The monitoring itself is organized cyclically.
In another study, the authors used cameras as a form of audiovisual commu-
nication with patients. Their goal was also to increase the level of protection of
medical staff, as well as to monitor the condition of patients from a remote location
[8]. In addition to the cameras, the use of the system itself required the installation
of a server, which was used to store the values of the measured parameters. The
group of vital patients parameters that were monitored by the proposed system
included the patient’s temperature, ECG, SpO2, blood pressure, CO2 and respira-
tory rate. The values of all parameters can be checked by the medical staff on a
mobile phone. In this way, the mobile phone is a special tool used to monitor the
patient’s condition, as well as to alert medical staff if the patient’s condition has
deteriorated. The key thing in the whole system is the fact that for their own safety,
the fashion staff dreams of contact with patients using such a system [9].
Authors in the one of the researches proposed a smart medical system for
monitoring both the patient’s condition and the condition of the room in which the
patients are. The system is based on the use of the IoT platform and a set of selected
sensors. The system was intended to monitor the condition of patients in hospital
[10–12]. The given system also monitored the situation in their rooms. A total of
five sensors were used to measure: heart beats, body temperature, room tempera-
ture, the amount of CO in the room, the amount of CO2 in the room [13]. The use of
such a system made it possible to monitor the condition of patients using a web
portal. In this way, the medical staff was able to analyze the patient’s condition at
any time, which increased the effectiveness of treatment. The error rate within the
system was below 5%.
In one of the researches, the authors created a mathematical model intended for
the analysis of future growth of epidemic. In the same time, this model needs to
predict how fast epidemic will grow. This model based on machine learning was
applied during 2020. The main task of the created prediction model was to assess
the speed of spread and danger of this virus in countries around the world [14]. The
entire process of creating, training and using the prediction model was performed
on the cloud platform in order to obtain more accurate results in real time. This
approach to data storage and processing is beneficial to both government and
citizens. Cloud computing within this model offers access capabilities regardless of
user location.
The research presented in [15] was conducted in two major directions. The first
direction was the challenges that the hospitality and tourism industry have due to
the new situation with the COVID-19 virus pandemic. The second line of research
concerned the vital industry. The research was based on interviews in which 15
participants participated. Participants were people in senior positions in hospitality
and tourism. By analyzing the results obtained after all the interviews, 27 subtopics
were created, which were later grouped into four larger groups. Out of the total
number of 27 subtopics, some dominant subtopics singled it out. The main goal of
the research was to determine the basic topics for combating a further decline in the
tourism and hospitality rate caused by the COVID-19 virus pandemic.
Author in [16] presents latency minimization in e-healthcare through fog
computing. More precisely, the authors presented a new scheme that uses machine
learning, more precisely the k-fold random forest algorithm, to segregate multi-
media data. The authors also presented models for calculating the total latency. By
the simulation procedure, they repented that it is possible to create a model that will
have a classification accuracy of 92%. The model also created showed a reduction
in latency of approximately 95% compared to previously existing models. In this
way, the quality of services belonging to e-health has been increased.
Authors in research presented in [17] show their model that should allow for
anonymity. This model is based on the use of IoT technologies. The authors pre-
sented the RFID concept by which it is possible to track nonhuman virus carriers.
Practically in this way, it will be possible to follow both animals and objects that
move like cars. By applying the model, animals and moving objects will be able to
send or receive notifications if they approach the marked area. The simulation
results showed very high efficiency.
Panic shopping at the beginning of each of the pandemics can cause great
logistics and supply to stores. The authors in one of the researches dealt with the
analysis of panic purchases during the corona virus pandemic [18].
The contribution of machine learning applied to data obtained during a pan-
demic can facilitate decisions about its further course. The aim of one of the papers
was to review the contribution of machine learning and IoT in the fight against the
COVID-19 epidemic [19].
8.3 Proposal of software solution for health protection

The main global focus of actions for almost all countries in the world in the last 2
years was how to prevent the spread of a pandemic, and how to minimize its effect
to different sectors such as economy, tourism, healthcare, and education. One of the
greatest challenges almost all countries faced was how to protect the economy,
when almost all industry stops working. As tourism, given the specificity that
characterizes it (travel, gatherings of people, social events, etc.), is largely affected
by the pandemic, it can be said that it has suffered the greatest losses compared to
other industries. The collapse of the economy coming from tourism has directly
affected a large number of families [20]. As the economies of almost all counties
worldwide rely directly on the revenues that are generated from the tourism sector,
it is necessary to find the best possible solution to relax the measures introduced in
response to the corona virus pandemic. Given the seriousness of this pandemic, the
opening of tourist destinations for tourists as well as the relaxation of the intro-
duced measures must be carried out in a way that will not endanger the health of
both natives and tourists.
Observed from the stakeholders’ point of view and principally having in mind
the current knowledge about the way the coronavirus spreads, as well as the ways to
fight against it, some of the key factors can be singled out. In the first place, special
attention should certainly be paid to tourists as potential carriers of the corona
virus. Viewed from the point of view of tourists as carriers, tourists can bring the
virus with them to the tourist destination where they are staying. Also, if, during
their stay at a specific tourist destination, they come into contact with the virus and
become infected, they can bring it home when they return and thus continue to
spread it. In this regard, a very important aspect in the protection of the health of
the population is certainly the precise testing and records of health institutions in
terms of tested and positive people for COVID-19.
Immunization of the population as one of the key measures in the fight against
the coronavirus opens up opportunities for easing measures, and restarting tourism,
as well as for facilitated travel and movement of tourists. This practically means
that it is necessary to establish a system that will provide facilities both in terms of
performing daily activities and in terms of enabling travel to the immunized part of
the population. Based on the regulations in force at the time of writing, persons who
have a certificate are not COVID-19 positive, as well as persons who have been
vaccinated, do not need to be in self-isolation. Each of the actors in the system has a
clearly defined role. Also, each of the actors in the proposed software system has a
predefined set of data with which it operates and for which it is responsible. The
system itself is organized as a series of services, grouped by levels. Mutual inter-
connection of all actors could be seen from Figure 8.1.
8.3.1 System architecture

The entire system is organized as a collection of different services, each one of
them providing different functionality. The architecture of this system is designed
to control access to sensitive data to various actors and provide valuable informa-
tion for each of them. Systemenables data travel toward the central storage system
and back to the actors. This component is organized as a set of services available to
other actors. The functionality of this component is based on four levels. These four
levels are named: core, support, analysis and application.
Core level may contain certain services like core data, commands, and meta-
data. Services on core level provide intermediary communications between the
actors of the system. The core data service family provides a centralized persistence
Application
level
Healthcare Local government Tourist

Border control
system system organization
Analysis
level
Analysis modules
Supporting
level
Logs Reports
Core level
Core data Command Database Metadata
Figure 8.1 Architecture of the proposed software solution
facility for data readings and also implements different protocols for data pro-
tections. If the data need to be preprocessed for clarification, or if there are
invalid instances that must be processed, this is where such a service will be
logically located. Command represents a service family that facilitates and con-
trols action requests from one side of the system to another. It should expose the
commands in a common, normalized way to simplify communications between
the actors. Metadata represent a repository and associated management service of
metadata about the actors that are connected to platform. Metadata provide the
possibility to create new sources, and store and manage information about the
services that serve as platform interfaces to the actual actors. Database services
are other services that communicate directly with the actors and normalize
information and communications.
The service layer provides support for the operation of software applications
such as logging, data cleaning, data scheduling. In our cases, supporting service
cannot be considered optional, as it is necessary to provide legal traces for data
sharing. These services can be fundamental because they could provide valuable
evidences in case of data leak or breach. Furthermore, this type of application
features can be very important for maintenance, bug tracing and further develop-
ment of the system. Service architecture can allow each of the actors to have their
own settings for logging and log file storage.
Since we created a system that has access to global data, it will be useful to
perform different analytics tasks on the collected and shared data. In this way, we
can get useful insights from the data that are relevant not only to public health, but
also for the entire human existence and well-being. Services in this level can use
data for which they have access permissions, making this entire system more
resilient for data breaches. Although this level is rather optional, we expect it to
become main focus of development in the future. With the increased data, the
demand for analysis will be brought to the fronts, because the knowledge extracted
from confirmed accurate data can be valuable for the entire population.
The role of application level is to extract, process/transform and send data from
platform to an actor. Services on the application level should be adjusted to meet
different requirements from different users. Furthermore, special attention must be
given to data privacy and data access control services. Each actor of the system will
be represented as a service on this level. This is the most efficient way to control
shared data and data access among the participants. Furthermore, services on this
level can use results and models created in analysis level to create different actions
or to generate comprehensive reports that can provide useful knowledge.
This way of organizing the management system was chosen to support
expansions while taking data privacy in mind. Each of the individual actors of the
system is enabled to operate on data on different devices. More about the individual
functionalities of each part of the system from the point of view of individual actors
will be given in the following sections. Each component on the application level
will be explained. Components on the other levels in this system are very general
and not specific to this particular implementation and usage and thus are not
explained in detail in this chapter.
8.3.2 Healthcare service

The proposed system identifies the healthcare system of each of the countries as a
stakeholder. As we mentioned earlier, the role of the healthcare system is to create
a register of infected people and enter the personal data of each of the patients who
are positive for the COVID-19 virus. These data are crucial in the fight against this
virus [21]. The healthcare service would have access only to the data for indivi-
duals who were tested positive to the virus. In this way, unnecessary sharing of
patients’ personal data will be reduced. Other actors need data on persons currently
positive for the COVID-19 virus as a starting point for future use. Each of the actors
in the system uses the obtained data in order to control and reduce the possibility of
irresponsible movement and spread of COVID-19 virus by infected persons. Such
tourists, depending on the competent authorities who discover their movements in
the period in which they should be in isolation, will be processed in an adequate
manner.
Another part of the healthcare service is a communicator for the IoT system
intended for monitoring the health condition of patients who are being treated
within a tourist destination. Practically, if tourists are hospitalized during their stay,
the system enables the collection of parameters such as body temperature, CO2
emissions and storage of all measured values on the cloud via a central software
platform. In this way, after they recover and return home, the local health institu-
tion has data on the patient’s condition and all the treatments he has undergone.
Data of this type can be entered within the system either by healthcare profes-
sionals, if their additional processing is required before the entry or automatically
using a data logger after each reading from the installed sensors. The role of the
data adapter as well as the filters shown in Figure 8.1, which belong to the
healthcare system, is reflected in the processing of data about each patient. In this
way, the recorded data on each individual patient required for the treatment of the
patient are processed so that only those data that are relevant will be forwarded to
the central system. The use of the mentioned adapter and filter prevents the
exchange of irrelevant data on patients of a specific healthcare institution. For
example, data on patients not recorded as COVID-19 positive would be considered
irrelevant data. Also if a person is recorded as COVID-19 positive, data on his or
her previous medical history will not be relevant within the proposed system and
should therefore certainly not be taken from the system of the healthcare institution.
Operating exclusively on approved data, as well as their targeted use by specific
system actors, reduces the possibility of their misuse.
Regarding the processing of data on the immunized part of the population,
depending on the immunization policy and recognition of vaccines of individual
manufacturers, each of the holdings within the system will define the rules
according to which other system actors will be guided when checking a particular
tourist. The task of the healthcare system within the immunization process is to
accurately record data on each person who received both the first and second doses
of the vaccine. Observed from the angle of the proposed software system, the
healthcare institution records data on each individual patient after vaccination.
After receiving the second dose, the healthcare institution should, in accordance
with the regulations issued by the state authorities of the country to which it
belongs, issue a certificate of vaccination to the patient. It should also add the
electronic data from the certificate to the central software platform of the proposed
software system. Electronic data on each vaccinated citizen in this way would
become available to all other actors in the system, which would enable easier
control of potential tourists, as well as tourists whose period of stay has
already begun.
8.3.3 Tourist service

Travel agencies as one actor of the proposed system have the task of entering data
on each individual reservation in the central register. In this way, anyone who
makes a reservation for accommodation or tourist tours can be checked through the
system by looking for matching personal data on the cloud. The task is to find
potential matches. Match detection refers to the search of data on the traveler who
wants to book accommodation, travel tickets, tourist tour and the like within the
proposed system. The first check that is made is whether there is a person within
the system who registers it as COVID-19 positive. If the person is COVID-19
positive or if he has been in contact with positive people, the reservation will not be
approved. Exceptions are cases when a COVID-19 positive person creates a
reservation for the period of stay that will be realized after the anticipated period of
isolation. Potential tourists who submit a negative PCR test or vaccination certifi-
cate are checked within the proposed system. Practically in this way, the authen-
ticity of the attached document is checked. If a person is indeed negative for
COVID-19 or has been immunized with one of the vaccines, it must be added to the
system by a healthcare institution that has performed either testing or immuniza-
tion. The employee within the travel agency forwards the data for verification
within the proposed system and receives a confirmation of the authenticity of the
documents. Vaccinated tourists must submit the green certificate or QR code in
order to prove that they received both doses of the vaccine. The speed of checking
the data of each of the tourists largely depends on the uniformity of the document
on the performed vaccination, which is issued by the authorized health institution.
This is especially reflected in cases when it comes to checking data for tourists who
are residents of another country or have not been vaccinated in their own country.
The application of such a system reduces the pressure on hoteliers, tourist
organizations as well as staff employed within tourist destinations. If each indivi-
dual tourist who paid for the stay through the agency has previously checked the
task of local tourist organizations, the hotel staff is reflected only in checking
tourists who have organized specific activities within the tourist destination
(accommodation, tourist tours, transportation, etc.) using various applications or in
person at the tourist place.
8.3.4 Local government service

As tourists can independently organize their stay in a tourist destination, the task of
local self-government is reflected in the entry and verification of data for each
tourist when registered by hoteliers. Practically, these are tourists who did not book
their stay through a travel agency, so they found accommodation upon arrival in the
desired tourist place. In order to take advantage of the services of a specific
accommodation, they must submit either a certificate of negative PCR test results
or a vaccination certificate. On the other hand, every registered hotelier must reg-
ister his guests in the branch office of the local tourist organization that operates
within the local government of the municipality in which they are located. The
tourist organization in cooperation with the local government institutions can check
the fulfillment of the conditions of stay of each individual guest. When registering
its guests and based on checking, the match of personal data of the guest with the
data available in the database of the proposed software system, the local tourist
organization using the software system received information whether a particular
person registered as COVID-19 is positive. In order to successfully perform the
test, they must have appropriate access to the system. By entering the data
belonging to a specific guest, as well as scanning with the reader both personal
documents and QR code on the certificate of negative PCR test, or certificate of
vaccination, the process of checking all documents begins. Also, the appropriate
institutions of local self-government are informed if the check shows that the
vaccination certificate has been falsified. Such an organized check once again
shows that uniformity is needed when it comes to documents confirming the per-
formed vaccination or negative results of the PCR test.
8.3.5 Border control

Border control between countries where there are border crossings can be said to be
the first barrier on which the separation of positive from nonpositive passengers can
be made. The primary goal of such screening is to keep positive individuals within
their country and prevent the spread of coronavirus. By doing this, it is possible to
contain the virus on geographic basis and to track its spread across the countries.
Border control needs to enter the data into the proposed system of all their
citizens who enter or left the country. The task of border control is reflected in the
fact that when checking the identity of a person entering or leaving the country, he
also checks whether a specific person is in the database of the proposed system, and
whether that person is registered as COVID-19 positive. Verification is used by
accessing the proposed software platform of the proposed system. If the passenger
is not in the system as a person whose test results have shown that he is positive for
COVID-19, the authenticity of the vaccination document is checked if the person
has been vaccinated and if he has the given document. Also, if the person is not
registered as COVID-19 positive, it is checked whether he has a certificate of
negative PCR test, the authenticity of which should be confirmed. In case when
passenger met all the conditions for entering in the country, the same will be
approved. Another functionality of the proposed system by the border control is the
entry of data on persons who were randomly tested when crossing the border. If
such persons are found to have tested positive at the border to be COVID-19, they
will notify the local self-government to which the place where the specific tourists
stayed is in order to locate them and quarantine them. Applying this principle
reduces the possibility of crossing borders for those people who are positive and
thus reduces the potential for the spread of the virus.
8.4 Data protection

The COVID-19 virus pandemic has undoubtedly affected daily activities in peo-
ple’s lives and work. Normal activities quite normal for the human race have
become something that must be avoided. On the other hand, something that is
personal and that should not be important to others has become a topic of the
majority. Among other things, the pandemic did not bypass human rights and
personal data protection. Many individuals and various organizations, under the
pretext of preventing the spread of the pandemic and its consequences, have given
themselves the right to access and publicly disclose the personal data of persons
infected or who have died from the COVID-19 virus. Precisely for this reason, the
question has been asked about the data that relate to others, which can be used in
cases of epidemics and the protection of humanity, and for what purpose they can
be used. On the other hand, it is necessary to clearly emphasize what are the per-
sonal data that, regardless of the situation, no one would have the right to use
without the prior consent of the person to whom they relate [22]. From the point of
view of the healthcare system, it can be said that in the event of a pandemic, the
health system is the one that has the most sensitive data of people infected with the
virus. A number of studies have shown that data on infected people are crucial in
order to create opportunities for their publication to process them and draw con-
clusions that can reduce the pandemic and thus reduce the number of infected
people. Also, monitoring the condition of patients using digital devices that log
each read parameter is of undoubted importance to the scientific community in the
process of creating predictive models that would much more accurately predict the
future condition of patients. Observed from the point of view of patients, and based
on published experiences, it is concluded that in the case of the COVID-19 virus
pandemic, the publication of data on infected people led to alienation and distance
between people, which greatly affected the psyche of patients and their
environment.
This way of collecting and processing data obtained in digital form by the
health system can lead to the fact that a large amount of both personal and non-
personal data from the group of sensitive data can be found in the data. If such data
are not used exclusively for the purpose of suppressing the spread of the pandemic,
but are misused, then such a processing procedure can lead to a serious violation of
human rights.
A potential problem that may arise in the process of collecting, publishing and
processing personal data of patients as well as data on their health status is the use of
these data even after the end of the epidemic. More precisely, the problem that may
arise is the public availability of data, which can enable individuals or organizations
to use them even after the approval issued during the epidemic, has formally ceased
to be valid. Precisely for these reasons, it is necessary to unequivocally adopt laws
and bylaws that would regulate this area, as well as that would regulate cases in
which it is possible to make exceptions. It is of utmost importance to control the
access to these data, if possible only to machines, which will generate summarized
and anonymized results when possible. Regarding the healthcare system, it is advi-
sable to apply the principle of patient data confidentiality.
In response to the problems of digital personal data protection during the
COVID-19 pandemic, a European Union directive on the privacy of electronic
communications was adopted. Within the framework of this directive, it is expli-
citly defined that data obtained from any source for processing for the purpose of
combating the COVID-19 pandemic may be used only if they are anonymous.
More precisely, if any information that may indicate which person is in question
has been removed from the data itself. If it is not possible to make the data anon-
ymous by removing information about the person in question, each of the member
states should prescribe special regulations that will regulate how to act in such
situations. Transferred to medical data, by removing the personal data of the patient
or his ID, it is possible to use all other information related to the example for
diagnosis, measured values of various parameters and so on.
After the adoption of the directive, many countries of the European Union and
beyond began to apply it, especially in the field of collecting the location of people of
interest. For example, in Italy, mobile operators have made data on the locations of
their users available to government services. This was of particular importance to the
Ministry of Health as it made it possible to point out citizens who had disobeyed a
restraining order or home isolation. Similar to the Italian model, the Polish authorities
created a software solution that used mobile operators to wash citizens who should be
in quarantine. The application is created so that from time to time, without the exact
time interval, it requires citizens to take selfies with geolocation to determine whe-
ther the quarantine measure is adhered to or not. The authorities in China have gone
even further, so that every citizen, when returning from abroad, receives an electronic
bracelet that sends geolocation, as well as a warning to the authorities if there is a
violation of the self-isolation measure. Singapore, on the other hand, has made all
data on patients and citizens tested positive. This procedure is designed so that all
citizens who are in the database can be followed and in order to discover which part
of the country can be a potential hotbed.
Articles 6 and 9 of the GDPR define legal norms related to employers and
healthcare organizations, which regulate the handling of personal data in the epi-
demic. However, the exchange of information in such cases must be proportional to
the risk and the actual need for exchange, meaning that organizations must check
regularly and carefully if the data access is absolutely necessary. It is up to employers
to adopt the measurements that must be taken to protect their employees [23].
The protection of personal data was certainly one of the aspects when devel-
oping the proposed software solution. The use of data obtained from the healthcare
system within the proposed system is done with special attention aimed at their
protection [24].

The corona virus pandemic has visibly disrupted people’s daily routine, their habits
and life plans. In order to prevent the further spread of the corona virus, measures
have been introduced that restrict movement, travel, gatherings and everything that
makes human beings the way we are. The tourism sector has suffered heavy eco-
nomic losses by closing borders and banning the operation of accommodation
facilities within tourist destinations. The proposed software solution aims to pro-
vide tools to system users to work together to reduce the impact of the pandemic on
the work of the tourism sector and restart tourism. The main task is certainly to
provide a mechanism to protect the health of the population, in order to prevent the
re-expansion of the pandemic.
The proposed software solution is architecturally based on a service archi-
tecture to allow decoupling and expansion of the system. The service architecture
of the proposed system is designed in a way that is suitable for adaptation and
expansion. That is a very important property of a system that should be used for
public data, because once integrated, it would not be easily replaceable or
architecturally modified. The main task of this platform is to provide a mechanism

of communication among several actors of the system taking into account the
privacy of the data. As it works with data that belong to the group of personal data,
mechanisms for the protection of this data must be provided, and the system must
be developed with privacy in mind. By decoupling the entire system to different
services with strict data access rights, actors in the system access only the set of
data that is important to them. Also, if it is necessary to access a broader set of data,
each access is recorded and searches over the data are performed independently of
the system actors, while it presents the required results. Furthermore, it is possible
to run an extensive analysis on the data that can be configured to only show results
relevant to broad population and anonymize the data to hide personal information.
The use of such a system, both locally and globally, by as many countries in
the world as possible will enable the relaxation of measures to combat the cor-
onavirus, which further leads to greater travel opportunities and economic
strengthening of tourist destinations. The basic precondition for the successful
implementation of such a system, which would offer the possibility of relaxing
measures and strengthening the collapsed tourism sector and the global economy of
many countries, is the continuous and joint work of state and world institutions.
The problems caused by the COVID-19 virus pandemic are far from being
solved. As the pandemic affected almost all branches of the economy, the economic
consequences are still expected. In order to mitigate the economic consequences in
the tourism sector, it is almost immediately necessary to act and enable unhindered
travel and work of tourist destinations. On the other hand, given the fact that the
spread of the virus has not stopped and that new strains of the virus are emerging,
everything needs to be approached in the right way. The proposed solution requires
mutual coordination and work of both the tourism sector and government institu-
tions at the local and state levels. Further research will be focused on the full
implementation of the proposed system and its application in real conditions. It is
expected that the system can greatly contribute to preventing the spread of the virus
by reckless behavior of individuals. The advantages of using such a system will
enable its use in cases of other viruses, not only COVID-19.
Further continuation of the described research will include the implementation of
the test model of the proposed system. The implementation of the test model of this
system should show through a series of simulations and practical use what is the
success rate of the application of such a system. A high success rate is expected, which
would indicate the fact that the implementation of such a system in practice is justified.
References
[1] Ventrella E. ‘Privacy in emergency circumstances: data protection and the

Covid-19 pandemic’. ERA Forum. 2020; 21, pp. 379–393.
[2] Gössling S., Scott D. and Hall C. M. ‘Pandemics, tourism and global change:
a rapid assessment of Covid-19’. Journal of Sustainable Tourism. 2020; 29
(1), pp. 1–20. https://doi.org/10.1080/09669582.2020.1758708.
[3] Farzanegan M. R., Gholipour H. F., Feizi M., Nunko R. and Andargoli A. E.
‘International tourism and outbreak of coronavirus (Covid-19): a cross-
country analysis’. Journal of Travel Research. 2021; 60. https://doi.org/10.
1177/0047287520931593.
[4] Olutosin T. and Absalom E. E. ‘Smart healthcare support for remote patient
monitoring during covid-19 quarantine’. Informatics in Medicine Unlocked.
2020; 20, pp. 1–11.
[5] Geetanjali S., Munish S. and Ajita K. ‘Evolution of smart healthcare: tele-
medicine during COVID-19 pandemic’. Journal of the Institution of
Engineers (India): Series B. 2021, pp. 1–6.
[6] Mohammed N. A., Halimatuz Z., Sairah A. K., Rusyaizila R. and Eddy Y.
‘Novel covid-19 detection and diagnosis system using IoT based smart hel-
met’. International Journal of Psychosocial Rehabilitation. 2020; 24(7),
pp. 2296–2303.
[7] Naveen N. R. G., Ajay S., Shiv L. S. and Puri G. D. ‘Real-time smart patient
monitoring and assessment amid COVID-19 pandemic – an alternative
approach to remote monitoring’. Journal of Medical Systems. 2020; 44(131),
pp. 130–131.
[8] Singh A., Naik B. N., Soni S. L. and Puri G. D. ‘Real-time remote surveil-
lance of doffing during COVID-19 pandemic: enhancing safety of health
care workers’. Anesthesia & Analgesia. 2020; 131, pp. e112–e113.
[9] Zhang Y., Liu H., Su X., Jiang P. and Wei D. ‘Remote mobile health mon-
itoring system based on smart phone and browser/server structure’. Journal
of Healthcare Engineering. 2015; 6(4), pp. 717–737.
[10] Rahman A., Chakraborty C., Anwar A., et al. ‘SDN-IoT empowered intel-
ligent framework for Industry 4.0 applications during COVID-19 pandemic’.
Cluster Computing. 2021, pp. 1–18. 10.1007/s10586-021-03367-4.
[11] Dash S., Chakraborty C., Giri S. K., Pani S. K. and Frnda J. ‘BIFM: big-data
driven intelligent forecasting model for COVID-19’. IEEE Access. 2021; 9,
pp. 97505–97517. 10.1109/ACCESS.2021.3094658.
[12] Aslam B., Javed A. R., Chakraborty C., Nebhen J., Raqib S. and Rizwan M.
‘Blockchain and ANFIS empowered IoMT application for privacy preserved
contact tracing in COVID-19 pandemic’. Personal and Ubiquitous
Computing. 2021, pp. 1–17. 10.1007/s00779-021-01596-3.
[13] Islam Md. M., Rahaman A. and Islam Md. R. ‘Development of smart
healthcare monitoring system in IoT environment’. SN Computer Science.
2020; 1(185).
[14] Tuli S., Tuli S., Tuli R. and Gill S. S. ‘Predicting the growth and trend of
COVID-19 pandemic using machine learning and cloud computing’.
Internet of Things. 2020; 11.
[15] Kaushal V. and Srivastava S. ‘Hospitality and tourism industry amid
COVID-19 pandemic: perspectives on challenges and learnings from India’.
International Journal of Hospitality Management. 2021; 92, pp. 1–19.
[16] Kishor A., Chakraborty C. and Jeberson W. ‘A novel fog computing
approach for minimization of latency in healthcare using machine learning’.
International Journal of Interactive Multimedia and Artificial Intelligence.

2020, pp. 1–11. dx.doi.org/10.9781/ijimai.2020.12.004.
[17] Garg L., Chukwu E., Nasser N., Chakraborty C. and Garg G. ‘Anonymity
preserving IoT-based COVID-19 and other infectious disease contact tracing
model’. IEEE Access. 2020; 8, pp. 159402–159414.
[18] Roy S. and Chakraborty C. ‘Panic buying situation during COVID-19 global
pandemic’. Journal of Information Technology Management. 2021; 13(2),
pp. 231–44.
[19] Chakraborty C. and Naser Abougreen A. ‘Intelligent Internet of Things and
advanced machine learning techniques for COVID-19’. EAI Endorsed
Transactions on Pervasive Health and Technology. 2021; 7, pp. 1–14.
[20] Gretzel U., Fuchs M., Baggio R., et al. ‘e-Tourism beyond COVID-19: a call
for transformative research’. Information Technology & Tourism. 2020; 22,
pp. 187–203.
[21] Zwitter A. and Gstrein O. ‘Big data, privacy and COVID-19 – learning from
humanitarian expertise in data protection’. Journal of International
Humanitarian Action. 2020; 5(4), pp. 1–7.
[22] Brouder P. ‘Reset redux: possible evolutionary pathways towards the trans-
formation of tourism in a Covid-19 world’. Tourism Geographies an
International Journal of Tourism Space, Place and Environment. 2020;
pp. 1–7.
[23] Fahey R. A. and Hino A. ‘Covid-19, digital privacy, and the social limits on
data-focused public health responses’. International Journal of Information
Management. 2020; 55, pp. 1–5.
[24] Spalević Ž. and Ilić M. ‘Management of tourist personal data protection in
COVID 19 – it perspectives, legal and economic aspects’. 5th International
Thematic Monograph – Modern Management Tools and Economy of
Tourism Sector in Present Era. 2020. pp. 409–424.
Chapter 9
Analysis of Artificial Intelligence and Internet of
Things in biomedical imaging and sequential
data for COVID-19
Sinthia Roy Banerjee1, Saurav Mallik2, Tapas Si3,
Arijit Banerjee4, Shan Jiang5 and Sudip Podder6
Abstract
Biomedical data analysis is an exceedingly broad field. It includes array data ana-
lysis, biomedical image analysis, integrated or hybrid data analysis, and patient
data analysis using machine learning (ML) and artificial intelligence (AI). Array
data analysis can be further classified into RNA-sequence, single-cell RNA-
sequence, microarray, and ChiP-seq data analysis. Biomedical imaging encom-
passes different parameters like gathering of biomedical signal, formation of an
image, processing of an image, display of image, and the medical diagnosis that are
built on the features obtained from various images. AI was mainly used to break
down healthcare data and used to track and screen patients while Internet of Things
was used mainly for monitoring a patient remotely. There are different radiological
imaging processes that include the radiography, ultrasound, thermography, mag-
netic resonance imaging, nuclear medicine and computed tomography. We, in this
book chapter, provide a comprehensive survey (road map) on various array-based
sequence data analyses and biomedical imaging along with their integrated studies
for different tissue-specific dreadful diseases (such as cancer). We included the
integrated studies of biomedical imaging and array-based data analysis for the same
set of patients (samples) that covered the problem of combinatorial gene signature
detection as well as disease subtype image classifications while specific multi-
modal data from well-known data repository (e.g., TCGA, ICGC) had been
1
Department of Computer Science and Engineering, University of Calcutta, Kolkata, India
2
Center for Precision Health, School of Biomedical Informatics, University of Texas Health Science
Center at Houston, Houston, USA
3
Department of Computer Science and Engineering, Bankura Unnayani Institute of Engineering,
Bankura, India
4
Siemens Healthineers Pvt. Ltd., Bangalore, India
5
Pamela Sklar Division of Psychiatric Genomics, Icahn School of Medicine at Mount Sinai, New York,
USA
6
Institute for Integrated Circuits, Johannes Kepler University Linz, Linz, Austria
provided. Finally, our book chapter covers the maximum area of biomedical ima-
ging as well as array-based sequence data analysis along with the contribution of AI
and ML in order to build a smart healthcare system, and provide a new dimension
to the interested biomedical researchers.
Keywords: Biomedical imaging; Microarray; RNA-sequence (RNA-seq);

Single-cell RNA-sequence (scRNA-seq); De novo mutation (DNM); Chromatin
Immunoprecipitation Sequence (CHiP-Seq); Computed tomography (CT);
Magnetic resonance imaging (MRI); Nuclear medicine imaging (NMI)
9.1 Introduction
The anthology of biomedical information is a major important element of research

in the domain of both biological and social science. Both the collections and usage
of biomedical information in social research have been growing rapidly. This
increase reflects both the growing recognition of the importance of understanding
the relationships between biomedical measures of health and individual outcomes
and the growth of multidisciplinary research frameworks that blend social and
medical paradigms of knowledge. The recent outbreak of COVID-19 pandemic has
turned the world upside down. Biomedical data analysis helps in understanding the
risks and signs with different consequences detected in patients with COVID-19.
Image analysis, array analysis, and hybrid analysis play an equally important role in
identifying the deadly COVID-19. Array data analysis is the procedure of exposing
a DNA, RNA, or any peptide sequence to abroad range of analytical processes so
that it can understand its features, its functionality, its structure, and its evolution.
We have even observed that array is immediately competent to differentiate clean
from recuperative COVID-19 subjects. Various methodologies used here include
sequence alignment, search against different biological databases. Array data ana-
lysis can further be classified as RNA-seq, single-cell RNA-sequence (scRNA-seq),
microarray, and ChiP-seq data analysis. Moreover, biomedical imaging as stated by
Jiang [1] has been also undergoing a rapid technical advancement over the last
couple of decades. It has been observed the development of various new applica-
tions to biomedical imaging. Different procedures such as functional imaging,
imaging related to spectroscopy, optical imaging, and the various image-guided
interventional techniques related to treatment or therapy have been attaining the
recognition in different domains extending from fundamental research to clinical
applications, and also from the cellular to the tissue and even to the organ level.
Biomedical imaging is a multidisciplinary field that requires the efficient colla-
boration among researchers of different fields, including biologists, medical phy-
sicists, chemists, pharmacists, computer scientists, statistician, biomedical
engineers, and other clinicians of different specialties. Efficacy, skill, and instru-
ments provided by the different radiologists are an integral part of various clinical
research programs [2]. Application of biomedical imaging is rising day by day and
Analysis of Artificial Intelligence and Internet of Things 229
it is becoming an emerging research topic as an essential fundamental clinical

investigational pro forma. Interestingly, biomedical imaging has a key role in
cancer research. Array-based next generations sequencing data analysis is another
hot biomedical research topic in this current era [3,4]. DNA microarray [5] is a
structured and efficient technology that can simultaneously identify all levels of
hundreds and thousands of transcripts (viz., miRNAs [6], genes, lncRNAs) across
various sample groups or tissue samples [3,4,6,7]. Besides biomedical data analy-
sis, artificial intelligence (AI) in healthcare primarily implies to physicians and
medical service providers retrieving enormous datasets of hypothetically life-
saving data. This involves medication techniques and their results, endurance
levels, and pace of treatment collected around many patients, topographical sites,
and numerous and at times interrelated well-being situations. Innovative processing
capability can identify and evaluate huge and minor developments from the records
and yet produce projections via machine learning (ML) that is intended to recog-
nize prospective well-being results. Internet of Things (IoT)-supported machines
have made remote monitoring in the healthcare sector possible, making it able to
maintain patients secure and healthy, and enabling doctors to provide excellent
treatment. It has also improved patient commitment and contentment as colla-
borations with doctors have turned out to be simpler and more effective. Moreover,
remote monitoring of patient’s well-being supports in decreasing the duration of
hospital stay and avoids readmissions. IoT also has a key effect on lowering
healthcare expenses substantially and thereby enhancing medication results.
A lot of works have been performed on all of these, but we think lots are yet to
be done. Thereafter, there is a huge challenge for new researchers to grab these
domains together, along with their integrated study to have a smooth progress in
their research. Hence, in our book chapter, we portray a thorough survey or (road
map) on various biomedical imaging and array-based sequence data analyses along
with their integrated studies for different tissue-specific dreadful diseases (such as
cancer) and usage of smart technologies. In this regard, initially, we provided dif-
ferent algorithms on biomedical imaging [8] that included multilevel thresholding
[9,10], Markov random field (MRF) model [11], and clustering algorithms [12].
The advantages, shortcomings, and usages of different algorithms of biomedical
imaging are demonstrated here. Second, we yielded a different array-based
sequence data analysis algorithm that is very much useful for biomedical
research. These array-based data might be divided into different categories, viz.,
single-cell sequencing data, RNA-seq data [6], microarray data. Various algorithms
applied to each category of the sequence data have been notified here. This includes
various computational problems such as gene signature discovering, biomarker
discovery, disease classification or disease subtype classification, differential
expression or differential methylation analysis, prognosis study, gene module
detection [13,14], multi-modal data integration, etc. For each problem, specific
algorithms (or, tools) have been mentioned here for the benevolent of the new
researchers in this domain. The advantages, limitations, and overall usages of each
category of the algorithms along with the corresponding comparative study of
the different categories of algorithms are also described here. Furthermore, we
Biomedical
data analysis
Array data Integrated or

analysis hybrid data
analysis
Single-cell
Biomedical RNA Chip
RNA Microarray
image -sequence -sequence
-sequence data analysis
analysis data analysis Data analysis
Data analysis
Computed Magnetic Nuclear

Ultrasound tomography resonance medicine
(CT) Imaging(MRI) imaging
Figure 9.1 Classification of biomedical data analysis
included the integrated study of biomedical imaging [15,16] and array-based data
analysis for the same set of patients (samples) that covered the problem of com-
binatorial gene signature detection as well as disease subtype image classifications
while specific multimodal data from well-known data repository (e.g., TCGA,
ICGC) had been provided. Finally, our book chapter covers the maximum area of
biomedical imaging as well as array-based sequence data analysis along with the
contribution of AI and ML in order to build a smart healthcare system and provide a
new dimension [17] to the interested biomedical researchers. The chapter aims to
provide researchers a review on various topics of biomedical data and the
advancement of new technologies that can be used to assist in battling COVID-19
disease.
The major contributions of the book chapter include the following: the first
section demonstrates the definition of various biomedical data and a comparative
study on different computational algorithms (AI- and IoT-based techniques)
applied to the biomedical data, while the second section of this book chapter covers
the advent of smart technologies during COVID-19 [18,19]. The last chapter
describes the conclusion regarding the empowerment of the new researchers to
further work on evolving additional solutions to combat COVID-19 and its future
applications in other biomedical applications (Figure 9.1).
9.2 Definition of biomedical keywords
9.2.1 Microarray and RNA-seq data

Microarray is a useful tool applied to determine the expression levels of thousands
of transcriptomes (genes) in the same time. DNA microarrays are termed micro-
scope slides that are used to print with thousands of tiny spots in specific predefined
positions, while each spot consisting of a defined DNA sequence (gene). The
microarrays are easier to use and cost-effective. They do not require huge scale of
DNA sequencing.
RNA sequencing (RNA-seq) is a process that is used to study the magnitude
and structures of RNA in a sample employing the next generation sequencing
technique. We can get RNA-seq data from various sources like ICGC, TCGA,
NCBI Gene Omnibus as stated by Geng et al. [2]. scRNA-seq is a demanding topic
used since the last decade. The scRNA-seq data are a prevalent and robust tech-
nology that permits researchers to describe the entire transcriptome of a huge
quantity of specific cells. scRNA-seq data are a popular dataset nowadays, but
analyzing scRNA-seq data is difficult due to technical variability, high noise levels,
and its massive sample sizes. scRNA-seq data have more dropouts and most often
generate sparse matrix that makes it a challenging process to identify original zeros
and dropouts.
9.2.2 De novo mutation

De novo mutation, as stated by Jiang et al. [1], refers to non-inherited mutation or
most specifically a mutation that is not present in every cell of parents. So, it
implies that the mutation happened in the sperm or the egg that generates the
offspring. Basically, the genetic modification that occurred initially in one of the
family members as an outcome of a metamorphosis in a germ cell, i.e., either egg or
sperm of one of the parents, or any variation that develops in the fertilized egg by
own through premature embryogenesis is termed de novo mutation. This is further
detailed in Table 9.5.
9.2.3 ChiP-seq data

ChiP-seq stands for immunoprecipitation mixed with high-throughput sequencing.
It detects the locations of genome attached by chromatin binding. ChiP-seq is a
prevailing process to find the genome-wide DNA-binding sites for transcription
factors (TFs) and many other proteins by combining chromatin immunoprecipita-
tion (ChiP) with sequencing.
9.2.4 Biomedical imaging

Medical imaging is a procedure of generating visual interpretations of our interior
body that is used for clinical diagnosis and medical research as well as the visual
illustration of the function of few tissues or organs physiologically. Medical ima-
ging tries to expose the internal structures that are hidden by our skin and bones as
well as to detect and heal different diseases. Moreover, it also creates a database
comprising normal anatomy and physiology to detect deformities. The localized
measurement of a contrast generating biophysical effect in body or organ of living
system is termed bioimaging. Now this contrast is basically the ability to distin-
guish tissue features against noise. Various modalities of medical imaging include
X-rays, computed tomography (CT) imaging, ultrasound (US), magnetic resonance
imaging (MRI), and nuclear medicine imaging.
9.3 Categories of computational algorithms in

biomedical data
There are three kinds of data analysis available, viz., (i) algorithms for biomedical
imaging data analysis, (ii) algorithms for array-based data analysis, and (iii) algo-
rithms for hybrid data analysis containing same set of patients.
9.3.1 Biomedical data analysis

The biomedical data analysis plays an indispensable role in improving medical
treatment, by data mining and integrating health records to reveal patterns that can be
used to craft informed policy. Thereby learning the skills to help turns data into
knowledge that can revolutionize the delivery of healthcare is of great importance.
Biomedical analysis has multiple facets that help us in diagnosing ailing organs of
human body. Hereby, we would like to throw some light on the way biomedical data
analysis helps in detecting malignancy specifically in brain and breasts. The most
widespread and important reasons for the rise in the death among adults and children
around the whole world is brain tumor. Brain tumor is basically a cluster of unusual
cells that expand inside the brain or around the brain. Few brain tumors are non-
cancerous (benign), whereas few are cancerous (malignant). Now computer-aided
detection or diagnosis, i.e., CADe or CADx, systems can strengthen the diagnostic
potentiality of doctors and minimize the time period necessary for the correct diag-
nosis. In the year 2014, Dahshan et al. [7] proposed a technique called the hybrid
intelligent ML for CADe system meant for automatic recognition of brain tumor using
MRI. The aforesaid method is basically based on the following computational pro-
cedures like the feedback pulse-coupled neural network used for image segmentation,
then the discrete wavelet transform used for features extraction, then the principal
component analysis used for reducing the dimensionality of the wavelet coefficients,
and lastly the feed-forward back-propagation neural network to categorize the inputs
into normal or abnormal category. Another most popular category of cancer found in
women all around the globe is the deadly breast cancer. An initial discovery as well as
the analysis of breast malignancy may lower the death ratio to a significant degree.
Now to detect breast malignancy, various kinds of imaging modalities are being
employed like CT, MRI, mammography, US, and biopsy. However, the different
histopathological images are attained since the biopsy could impact the process cancer
that is detected and the phase it is currently in. The computer-assisted diagnosis
technique facilitates the diagnosticians in the early identification of the breast malig-
nancy. The survey as stated by Kaushal et al. [3], in recent times, described various
methods of the breast cancer analysis employing various histopathological images.
Breast MRI possesses maximum sensitivity for breast cancer recognition among
recent medical imaging modalities and is very crucial for the breast imaging practices.
The foundation of breast MRI mainly consists of T1-weighted contrast-enhanced
imaging, T2-weighted, ultrafast, and diffusion-weighted imaging out of which last
one, i.e., diffusion-weighted imaging may be utilized to develop lesion categorization,
as stated by Mann et al. [4].
9.3.2 Array-based data analysis

Array-based data analyses are used in deciphering the data generated from resear-
ches on DNA, on RNA, and on protein microarrays, which allows researchers to
investigate the expression, basically the status of a larger gene sets—in most cases,
or any organism’s entire genome in one single experiment. There are three types of
array data analysis, namely, microarray data analysis, the RNA-seq data analysis,
and the scRNA-seq data analysis.
9.3.2.1 Microarray data analysis

There are lots of algorithms that have been developed by researchers in the last few
years to identify the different gene modules. A new algorithm called WeCoMXP has
been developed to assess assimilating co-methylation, co-expression, and protein-to-
protein communications to identify modules of gene for multiomics dataset. The
algorithm computes the respective divergent values from the intended indistinguish-
able values and detects the remarkable gene-modules from the network that regulates
the gene using average linkage clustering. Subsequently, it looks out for literature and
KEGG pathway. Gene-ontology evaluates to authenticate the genes depicting the
modules. Moreover, it decides the TFs that control these gene-modules, and the
miRNAs that aims for these genes-modules. But the major drawback is that con-
nectivity measure for signed network is not taken into consideration at all. Then this
chapter by Chandrashekar et al. [5] throws a light on guaranteeing precise results with
minimal gene subsets allowing the doctors to identify the kind of datasets in cancer
disease. It signifies the same classifier utilized for the selection and classification of
gene, thereby improving the depth of the model portrayed by the results. Among
various algorithms and tools that are used in recent times, Moonlight stated by
Colaprico et al. [13] is a remarkable tool that uses multiomics data to recognize genes
driving critical cancer. Moonlight analyzes more than 8,000 tumors from 18 kinds of
cancer, thereby finding 3,310 oncogenic mediators and 151 possessing 2-fold roles.
Moonlight basically targets to detect biological methods enhanced by a set of differ-
entially expressed genes between couple of situations by means of functional
enrichment analysis to detect the uphill regulators of the biological techniques that are
significantly enhanced in contrast by means of upstream regulator analysis, and to
detect “driver genes” having a double role that works as TSG or OCG in multiple
stages or kinds of cancer. All the descriptions of the algorithms for microarray data
along with their advantages and limitations are highlighted in Table 9.3.
9.3.2.2 RNA-seq data analysis

In the year 2015, Maulik et al. [14] came up with their method named as the statistical
biclustering-based rule mining (StatBicRM) for detecting significant expressions of
gene and the reports of methylation data. The main advantage of this method is the
effectiveness of time consumption when dealing with bigger datasets. Then Mallik
and Zhao [15] came up with an innovative idea for identifying methylation signs in
various neurodegenerative diseases by using different systems that happen to use
density-based clustering and reducing noise from the method. Their analysis dis-
covered a methylation signature that comprises 21 genes for Alzheimer’s disease
commonly known as AD and a methylation sign containing 89 genes of down syn-

drome. Their assessment addressed the mentioned signs that can cause massive
categorization correctness numbers (92% and 70%) for the abovementioned disorders.
Precisely, this system will be beneficial as it improves detection outlier-free genetic as
well as the epigenetic signs in many complicated disorders and with its respective
progressive phases. Then in the year 2018, Bandyopadhyay and Mallik came up with
Combinatorial Marker Discovery to research on tumorigenesis [9]. They proposed a
unique computational framework, mainly used for detecting significant combinatorial
markers commonly termed SCMs by utilizing the methylation data and the gene
expression together. Both methylation data as well as gene expression remain com-
bined in sole contiguous data and (after being approximated) propositional data
founded on its respective inherent (i.e., reverse) affiliation. Thereafter, a collective
count of expression and methylation figures (viz., CoMEx) have been created and
calculated based upon continuous collaborative information for detecting an early set
of necessary genes. This particular technique generates a lesser number of important
nonredundant gene sets than produced by any different methods and it is compara-
tively quicker. However, the anticorrelation factor is not considered by them.
Elaborate discussion on RNA-seq data is given in Table 9.2.
9.3.2.3 scRNA-seq data analysis

In this current era, scRNA-seq has turned out to be widely used for transcriptome
analysis in numerous fields of biology. In the year 2017, Matsumoto et al. came up
with an effective regulatory network inference algorithm from the scRNA-seq
during their differentiation, i.e., SCODE as stated by Hirotaka et al. [10], which
they applied to three scRNA-seq datasets and computed SCODE value by evalu-
ating its implied associations using a DNA seq-footprint based network. SCODE
performed well almost for couple of datasets and average for the rest of the data-
sets. This algorithm efficiently concludes regulatory network that is based on
ordinary differential equation. This method fosters the growth of the single-cell
differentiation analyses and bioinformatics method. Since each scRNA-seq profile
represents a limited specimen of mRNA molecules from a distinctive cell that in no
way can be retested and needs a strong assessment; thereby, it must split the sam-
pling effect from the biological variance. Therefore, in 2019, Baran et al. [16]
illustrate an approach for dividing the scRNA-seq datasets into separate as well as
identical clusters of metacells that can be retested from the similar cell. This
algorithm, on the contrary to clustering analysis, also focuses on achieving micro
groups rather than the macro clusters. Here it portrays a way to utilize meta cells as
building blocks for any complicated mathematical transcriptional plans by averting
flattening of data. Now in the year 2021, Bhej et al. [8] establish an ML-based
oversampling technique that utilizes a resultant expression of genes from pre-
viously discovered unusual cells as an input in order to produce the artificial cells
to detect related unusual cells in other freely obtainable researches. They make the
use of single-cell synthetic oversampling that is again created from the Localized
Random Affine Shadowsampling algorithm. This algorithm rectifies the whole
inequality ratio of the smaller and the bulk category. It also demonstrates the
efficacy of the technique meant for a couple of individual use cases, every single
comprising a couple of available set of data. For point-wise discussion on scRNA-
seq data, see Table 9.1.
9.3.3 Hybrid data analysis

Hybrid data analyses have both imaging and array-based data for the same set of
patients. The present technological advance is taking place in developing brain ima-
ging data, across multiple modalities, as well as genomic data having high throughput.
The study of brain tomography is an evolving and promptly expanding research area.
It makes consolidative research that evaluates genetic differences like the single
nucleotide polymorphisms or SNPs, epigenetic and copy number variations known as
CNVs, the molecular features taken by the different omics data and the quantitative
traits (QTs) of brain imaging, combined with more biomarkers, medical as well as
ecological data. The main objective of genomics imaging is to obtain different per-
ceptions from the phenotypic traits and the genomic and molecular structures of
human brain, along with its effect on regular and disorderly brain functions as well as
its behaviors. Taking into consideration the most exceptional magnitude and intricacy
of the datasets having brain imaging genomics, most crucial computational and sta-
tistical trials have to be satisfied in order to achieve the complete ability of these
useful data. Now surmounting these issues has turn out to be a foremost and
demanding topic of study in the field of statistics and ML, where efficient and cost-
effective data analytic techniques are used to disclose the genetic and the molecular
underpinnings of neurobiological procedures, which may influence the advancement
of the investigative, therapeutic, and the precautionary approaches for complicated
brain syndromes. Shen and Thompson [6] have evaluated three main classifications of
learning different problems in the genomics of brain imaging. First, it concentrated on
understanding the problem of heritability approximation of the brain imaging QTs.
The heritability of characteristic is mainly defended by the genetic factor. Second, it
stresses on the issue of discovering the imaging genomics associations, an important
topic explored in the brain imaging genomics to obtain innovative perceptions into the
genomic and the molecular systems of the brain formation along with its purpose.
Third, it concentrated on the knowledge problem of image integration and genomics
for the outcome projection. This is a very vital issue studied related to brain imaging
to attain significant perceptions onto the consequence-related neurobiological systems
at the genetic, the molecular, and the macroscale brain structure degrees. Elaborate
discussion on hybrid biomedical data analysis having biomedical data and array data
together are given in Table 9.6.
9.4 Different techniques for diagnosis using biomedical

imaging
9.4.1 Brain
Brain tumor is the extremely suffered disease that requires early and accurate
detection techniques. Now most of the detection and diagnosis methodologies
depend on the decision of neurospecialists as well as radiologists for the image

assessment that leads to human inaccuracies and is also very time taking. Now
CADx is basically ML and other computer algorithms that guide doctors to
figure out what is going wrong in particular radiology images and other related
areas. The CAD system improves investigative competencies of doctors and
minimizes the span needed to analysis appropriately. The research paper as stated
by Dahshan et al. [7] reveals the current available segmentation, categorization
methods, and their advanced progression for human brain. The paper discusses the
current segmentation as well as categorization methods and their existing advanced
MRI images of human brain. On this context, this recommends a hybrid intelligent
ML method based on CAD system used for detecting brain tumor automatically
using MRI. The different computational techniques that have been proposed in this
method for the image segmentation for reducing the dimensionality of the wavelet
coefficients, for features extraction, are the feedback pulse-coupled neural network,
the discrete wavelet transformation, and the principal component analysis, respec-
tively, in order to classify inputs into normal or abnormal.
9.4.2 Breast
The usual type of cancer in the females across the globe is the deadly breast cancer.
But the early discovery and identification of the breast cancer can minimize the
mortality rate significantly. The various categories of imaging modalities that have
been employed in order to detect and accumulate samples are mammography, US,
CT, MRI, and many others. The CADx generally facilitates the doctors and
pathologists in the detection of breast cancer. Kaushal et al. [3] report various
methodologies for diagnosing breast cancer using various histopathological images.
The important steps included in CAD analysis are preprocessing, segmentation,
feature extraction, and classification. The digital image preprocessing is performed
soon after the image is acquired. This plays the most essential part in detecting the
organic tissues depicted in the image by enhancing its quality without obliterating
the vital aspects. The second most essential stage is to detect the cells and the nuclei
of various histopathological images that form the basis for the classification are
called the segmentation stage. The third step, i.e., feature extraction phase, gen-
erally indicates the idea of getting prominent features that are immutable to the
inappropriate changes of the input. Following the image segmentation stage, mostly
the mining of features is done at the tissue level or could as well be done at the
cellular level in order to quantify the changes. Finally, classification is used to
classify which new instance belongs to which set of categories based on the
existing dataset. For detecting abnormality in the images, different classifiers are
used to classify tissues in various groups on the basis of the various kinds or grades
of the breast malignancy.
9.4.3 Kidney
Another most common form of cancer is renal cancer as reported by the National
Cancer Registry Program and Surveillance Epidemiology, and this renal cancer
accounts for almost 3–4 percent of the entire fresh cancer incidents in Indian sub-
continent and in the Western part of the world. We know that CADx systems can
help enhance the abilities of physicians and also reduce the total span needed to
conduct the exact analysis. Only way to survive for patients suffering from kidney
cancer is premature discovery and therapy employing various imaging modalities
that are noninvasive. Currently, renal cancer tomography mainly depends on CT. It
basically is believed as the benchmark for identifying and categorization of kidney
nodules. Kaur et al. [11] throw light on the massive role played by CADx for
kidney nodules. It starts with an acquisition of image, preprocessing, segmentation,
extraction of feature, selection, and categorization. Chaudhuri et al. [12] suggested
another deep learning model known as U-Net for semantic segmentation of kidney.
The main reason for selecting U-Net is that it is suitable for a smaller set of data,
and it was initially devised for the process of biomedical image segmentation.
9.4.4 Ovary
Ovarian cancer is another most suffered disease. It can sneak up on us like a beast
at night veiled as ordinary problems such prolonged bloating, then fatigue and the
constant need to pee. The extremely common form of ovarian cancer initiates in
aphelion cells, a particular type of cell that basically creates the tissue covering up
the ovarian surface. To treat the ovarian cancers in the best possible way, doctors
first confirm the identification and then discover the stage or the extent of the
cancer that requires a technique called the surgical staging during which the doctor
investigates the pelvic and the abdominal areas. The stage of the ovarian cancer is
determined by the position of the cancer cells prevalent in our body. In the first
stage, it is generally found on the upper surface of one or both ovaries or in the
fallopian tube or in the abdominal fluid. In the second stage, the cancer gets pro-
pagated from the initial site to other nearby areas such as the ovaries, the fallopian
tube, or even the uterus. In the third stage, cancer gets spread outside the pelvis like
the lymph nodes or also on the surface of liver or spleen as stated by Togashi [17].
All modalities like CT, US, and MR imaging have almost equal precision rate when
it comes to identifying stage of ovarian cancer.
9.4.5 Skin cancer

Skin cancer is any type of malignant growth that happens in skin. It can be broadly
divided into two main categories that are melanoma and nonmelanoma skin cancer.
Nonmelanoma skin cancer can further be categorized to basal cell cancer and
squamous cell cancer. Most often basal cell cancer is diagnosed in any organ
worldwide. Therefore, it is more common than the breast cancer, lung cancer, or
even colon cancer. The second most common cancer diagnosed worldwide is
squamous cell cancer. Mostly, patients will have either the basal cell cancer or the
squamous cell cancer. Till date, biopsy technique is deemed as the benchmark used
for detecting skin cancer in medical field. Although there are various anatomical
imaging techniques available to assess the diverse kinds of skin cancer nodules,
which include laser scanning, confocal microscopy, optical coherence tomography,
high-frequency US, terahertz-pulsed imaging, MRI, and many more advanced

procedures like the photoacoustic microscopy as stated by Hong and Cai [20], but
the anatomical tomography may be inadequate in managing skin cancer detection
as well as treatment. So, a variety of molecular imaging methods (like single-
photon emission CT and the PET) are used to scrutinize the skin cancer imaging.
9.4.6 Soft tissue sarcoma

Soft tissue sarcoma is a sporadic type of malignant tumor of soft tissues at either of
the extremities like arms or legs or in the central part of the body as well. They
usually appear in the areas such as our muscle or in fat though they can occur in any
other soft tissues also. Among the different types of sarcoma, probably liposarcoma
is most frequently found. Mostly all soft tissue sarcoma is treated with surgeries
where the doctors take out the tumor with a narrow zone of tissues around it to
ensure the entire tumor and all of its cells are removed. Depending on multiple
clinical factors like patients age or aggressiveness of the tumor addition of radiation
or chemotherapy can also be followed. As stated by Afonso and Mascarenhas,
Reference [21] throws a light toward the rational diagnostic tomography technique
to patients having soft tissue tumors, underlining the basic principle that is an
integral part of the soft tissue tumor imaging and diagnosis. In patients suspected
with soft tissue lumps, US is well suited as the preliminary triage of the imaging
modality, as the alleged tumor is available by using sonography though deeply
seated tumors create apparent problems. Moreover, it is also a first-line analysis for
children. MRI method provides finest contrast of soft tissue, has multiplanar cap-
ability, but is deficient in ionizing radiation and hence has become the most chosen
modality to assess soft tissue masses. The CAD technology is another advanced
digitized histopathology. Elaborate discussion on different modalities for biome-
dical imaging of various organs is given in Table 9.4.
9.5 Comparative review of computational algorithms
The different categories of computational algorithms have been reviewed and a

comparative study is presented, as shown in Tables 9.1–9.6.
9.6 Role of CT in COVID-19 pandemic

We all know that in the month of December 2019, a tremendous epidemic of a
novel infection was reported in China at the city of Wuhan, known as coronavirus
or severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) as its origin.
The international public health institution, World Health Organization (WHO), had
thereafter identified the infection triggered by the novel coronavirus as coronavirus
disease 2019 (COVID-19). Thereafter on March 11, 2020, WHO announced
COVID-19 a pandemic, with approximately more than 10 lakhs confirmed cases
and almost 5 lakhs deaths globally, within June 2020. Now, reverse-transcription–
Table 9.1 Different algorithms for single-cell RNA-sequencing (scRNA-seq) data
S. Name of the method Category Description Advantages Limitations

no.
1 Cell clustering [22] Clustering of single A fuzzy clustering technique based on Used to group the cells Other biological
cells multi-objective optimization challenges are not
(MOO) was built to identify the cell considered here
clusters in scRNA-seq data
2. Comparative study on Dimension reduction Various methods like PCA are com- Data points are well distant More aspects need
dimension reduction and clustering stra- monly used for unsupervised di- when the dissimilar data to be evaluated
and clustering tech- tegies mensionality reduction method points have been concen-
niques [23] trated in the low-down di-
mensional area
3. Meta as stated by Graph partitioning To partition multimodal scRNA-seq ● Used to partition the graph
Barang cell [63] using KNN graph data profiles into several disjoint ● Building blocks for compli-
clusters of data profiles cated quantitative
transcriptional maps
4. Clustering and classi- Clustering scRNA-seq Discussed ML tactics for grouping Reviews nonsegregated meth- Provides expression
fication algorithms data and categorization of scRNA-seq ods and tools profile informa-
scRNA-seq data as and stresses on DR tion of single-cell
stated by Ren et al. data
[24]
5. SCODE as stated by Regulatory network Suggests an algorithm for scRNA-seq Efficiently concludes regula-
Hirotaka et al. [10] interference algo- data, carried out on differentiating tory network based on or-
rithm cells to revamp expression dynamics dinary differential equations
and concludes regulatory networks
6. Automated annotation ML-based oversam- Mentioned technique considers al- ● Rectifies the entire ● Time-consuming
as stated by Bej pling method ready identified rare cells and their imbalance ratio ● Requires domain
et al. [8] gene expression count as input to ● Recognizes unusual cell expertise
produce synthetic cells, therefore population with minimal knowledge
identifying similar rare cells false positive detection
(Continues)

no.
7. Computational data Data analysis of Provides an outline of scRNA tech- ● Ruling out cells that are ● Available
analysis as stated by scRNA-seq data nologies, demonstrating the proce- inferior in quality scRNA-seq
Geng et al. [2] dures for varied data of scRNA-seq methods dropout,
analysis inadequate gene
expression
ignored
8. Relevant datasets and Clustering of scRNA- scRNA-seq clustering models com- Improves the interpretation of Lacking the gold
analytical tools as seq data prising seven grouping methodolo- treatment effectiveness standard bench-
stated by Peng et al. gies are assessed on five public mark datasets
[25] datasets
9. scRNA-seq analytical Clustering and compu- Recommends the combined advance- ● The computational techni-
pipelines and com- tational methods ment of strong analytic pipelines, que findings are bench
putational methods deals the challenges offered by marked and potential pro-
as stated by Hie investigative techniques, appropri- spects identified
et al. [26] ate for genomic queries
10. Pathway analysis as Benchmarking algo- Broad framework to evaluate correct- Selects appropriate methodol-
stated by Zhang rithm for the path- ness, stability, and the scalability of ogies that efficiently and
et al. [27] way activity the several pathway activity trans- accurately analyze scRNA-
transformation of formation algorithm seq data
the scRNA-seq data
Table 9.2 Different algorithms for RNA-seq data

no.
1 StatBicRM as stated by Clustering of Computational framework for Execution time is saved and
Maulik et al. [14] genes and sam- rule mining StatBicRM (i.e., works on large datasets
ples statistical biclustering-based
rule mining) utilizing cohe-
sive methods of statistical as
well as binary biclustering
methods from the biological
datasets
2. Density-based spatial Hierarchical clus- DBSCAN algorithm is applied It is useful to detect outlier- Gene expression, copy number
clustering of applica- tering to get rid of outliers; Limma free genetic and epige- variation, chromatin remo-
tions with noise statistical method is used to netic signatures in several deling, etc. are not part of the
(DBSCAN) as stated establish differentially complicated disorders and study
by Mallik and Zhao methylated genes, and hier- respective evolving
[15] archical clustering method phases
was applied to identify gene
modules
3. H19-mediated multimo- miRNA-mediated It established a unique analytic ● The effect of coregulation
lecular network ana- lncRNA, TF, method by combining the of multi-molecules can be
lysis in pan cancer Li gene coregula- synergistic rule amongst determined
et al. [28] tion is complex transcription factors (TFs), ● Effect of H19 on the bio-
but crucial in lncRNAs (e.g., H19), micro- molecules can be derived
cancer RNAs (miRNAs), and target
genes and then used it across
datasets having cancer ex-
pression using linear regres-
sion from the Cancer Genome
Atlas
(Continues)

no.
4. Pareto-optimal cluster Cluster algorithm It introduced a fresh structure It is valuable to discover
[29] for detecting signatures of signature for any micro-
gene employing Pareto- array or RNA-seq data
optimal cluster volume de-
tection for RNA-seq data
5. Hypograph mining for Hypograph It statistically identifies signifi- ● Finds out the strength of Gene expression, methylation,
obtaining the multi- mining cant genes from each of the co-similarity in multiview copynumber, etc. which are
view cosimilarity data profile and thereby cal- gene modules used to evaluate the inherent
gene modules as sta- culates the set of union con- ● Proposed a theoretic biological relations among
ted by Bhadra et al. sisting of all the statistically information profiles are not included in
[30] meaningful genes naming MiNoMi this chapter
6. mRNA to miRNA reg- Identification of It explored a panel of miRNAs Used for the detection of Here subtype-specific miRNAs
ulatory network ana- miRNA panel and meager molecular sub- risky patients and prob- and their predictive implica-
lysis as stated by types in PDAC able disorder monitoring tions remain unknown
Kandimallaand et al. in the patients with PDAC
[31]
7. Combinatorial Marker Gene expression Combines the score of gene ● Method generates a smal- Strength of anticorrelation is not
discovery [9] and methyla- expression and methylation ler number of non- undertaken in this chapter
tion data data to compute the combined redundant gene sets
continuous data for discover- ● Faster
ing essential bunch of genes
8. Combined framework DNA methylation Determines predicted gene ex- Helps in better interpretation The approach is performed on
for accurate biologi- and gene expression number from the of DNA methylation only a small dataset
cal interpretation of pression already filtered methylation study and gene expression
DNA methylation, as information during linear re- in any disorder survey
stated by Mallik et al. gression by empirical Bayes
[29] test using Limma
9. Seminoma and non- Subtype classifi- Extensively investigates the pair Identifies the different reg- Here dataset does not include
seminoma in testicu- cation of TF and miRNA regulation, ulatory particles, its inter- matched control samples
lar germ cell tumors their FFLs, and TF-miRNA- action modules, along
as stated by Qin et al. mediated regulatory networks with few other character-
[32] in two kinds of testicular istics of TGCT subtypes
germ cell tumors (TGCT):
seminoma (SE) and non-
seminoma (NSE)
10. Identification of co- DNA methylation It is a novel consolidative study Method is highly useful to
methylated and co- of a framework to detect the obtain potential gene/
express genes as sta- co-methylated as well as the miRNA modules as well
ted by Mallik et al. coexpress genes like mRNAs as hubs for any kind of
[29] and microRNAs (miRNAs) diseases
modules in the TGCT sub-
types, for example, non-
seminoma and seminoma
Table 9.3 Different algorithms for microarray data

no.
1 Comparative study among dif- Supervised Supervised DMR findings under 60 Figure out the It covers only four supervised
●
ferent DMR finding algo- and unsu- different parameter settings are DMR finding DMR finding tool, rather than
rithms, as stated by Mallik pervised used, which computes the tool, will be describing any other unsu-
et al. [33] consecutive strength and accuracy region. useful for find- pervised tool
clustering Compares results of many more ing which
algorithm features of analysis that includescondition
the volume of the DMRs over- ● Find optimal
lapping the approaches and the parameter set-
performance time ting of each
DMR
finding tool
2. Comprehensive survey on dif- Microarray A thorough survey of several para- Useful for data Variability is ignored
ferent parametric and non- data metric and nonparametric ap- with very few
parametric testing proaches for finding samples
methodologies as stated by differentially expressed tran-
Bandyopadhyay et al. [34] scripts
3. Differential methylated regions DMR analysis coMethDMR is a flexible and The subregion
as stated by Gomezand et al. of array- powerful tool for identifying identification
[35] based DNA DMRs. coMethDMR selects co- phase in co-
methylation methylated subregions and tests MethDMR en-
data association between methylation hances power
levels within the subregion and significantly
the phenotype
4. Regulatory analytical approach Gene regula- Here the regulatory mechanism is Efficient Time-consuming
to detect the preserved regu- tory me- investigated with an analytic
latory networks in the human chanism methodology to recognize the
beings and the mice, as stated preserved supervisory links in
by Aimin et al. [36] human beings and rats to
detect critical miRNAs, target
genes and the regulatory mo-
tifs (miRNA-TF-gene)
5. Moonlight as stated by Colapri- Detection of Analyzes more than 8 000 sam- It explains tumor
coand et al. [13] cancer dri- ples of tumors from 18 kinds heterogeneity and
ver genes of cancer, uncovering ap- helps in therapeu-
proximately 3 310 oncogenic tic decisions
negotiators, having 151 2-fold
roles. Confirms various genes
driving critical cancers by
studying the cell-line datasets
6. WeCoMXP as stated by Mallik Gene module Develops a new weighted connec- It has got higher WeCoMXP operates for only
and Bandyopadhyay [37] detection tivity measure incorporating the performance in unsigned/undirected graph
comethylation, co-expression terms of nu- where the anticorrelation for
and the protein–protein interac- merous cluster- the co-expression verses co-
tions (called WeCoMXP) to ing validity methylation is not contem-
identify the gene-modules for the indices plated
multiomics dataset
7. Distributed parallel algorithm as A multi- Proposes a multi-objective feature Highly effective
stated by Cao et al. [38] objective selection model that concurrently and efficient
feature se- considers the classification error,
lection the feature number, and the
model feature redundancy
(Continues)

no.
8. Distributed correlation based Biomarkers Newfound technique based on filter ● High classifica-
gene selection as suggested by detection for selection of gene is presented. tion accuracy
Shukla and Tripathi [39] from mi- This picks appropriate genes to ● High
croarray separate the tissues from the performance
dataset having gene expression. ● Lower
This calculates the relation be- execution time
tween the gene–gene and the
gene–class to identify subset of
essential genes
9. Disease prediction using deep Microarray Evaluates the assessment period, Accuracy rate is
learning technique, as stated classifica- the categorization, correctness very high by
by Chandrasekhar et al. [5] tion and also the potentiality to detect using classifi-
the sickness to find out the cation and fea-
stringency positions of illness ture selection
and the empirical outputs, which
signifies the deep neural network
categorization execution
10. Heuristic algorithm as stated by Microarray Here a multi-objective heuristic Various subsets of High computational cost
Jia et al. [40] data classi- algorithm named MOEDA is genes get high
fication proposed according to analytic classification
hierarchy process (AHP) accuracies
Table 9.4 Different modalities for biomedical imaging of various organs
S. Name of var- Modality used Advantages Disadvantages

no. ious organs
1. Brain [7] ● CT CT acquires detailed images of bone, soft tissue, ● CT involves radiation exposure
● MRI blood vessels, and various kinds of tissuesMRI ● Artifact from metal implants will conceal the
offers better soft tissues comparatively and dif- view in CT images
ferentiate better between the fat, water, and other
● MRI scanners may cause a safety issue due to its
soft tissues
strong magnets
2. Breast as stated ● Mammography ● Lessens the probability of undergoing ● Mammograms expose the breasts to radiation
by Mann ● US chemotherapy over time
et al. [4] ● US gives a clear picture of soft tissues and helps ● US cannot screen breast cancer as it misses
● MRI
detecting lesions in women having dense breast early symptoms because it can only detect
● MRI delineates the true size of a cancer that often whether cyst is filled with fluid or a solid tumor
happens to be underestimated on ● MRI cannot distinguish between cancerous
mammography and US abnormalities that lead to unnecessary biopsies
known as false-positive results
3. Kidney as stated ● US ● US modality helps in finding a kidney mass and ● US overestimates kidney stone sizes
by Nikken ● CT shows whether it is a solid or a mass filled with ● CT characterizes hypo-attenuation in masses
and Krestin a fluid generally smaller than 8–10 mm
[41] ● MRI
● CT helps us to identify if the kidney has tumor or ● In MRI, kidney seems as fuzzy signal voids,
lesions or it has been through disruptive condi- hence, gets overlooked with any other structures
tions like small stones, genetic variances, poly- or artifacts
cystic ailments, accumulation of water near
kidney, or around abscesses
MRI detects better contrast of soft tissue, thus iden-
tifying and categorizing renal lesions
(Continues)
S. Name of var- Modality used Advantages Disadvantages

no. ious organs
4. Ovary as stated ● TVS ● TVS acquires effective images that help differ- ● TVS is unreliable in differentiating benign from
by Togashi ● CT entiating simple cysts from complicated cysts malignant ovarian tumors
[17] ● CECT studies can compare to lower dose of non- ● CT scans do not show small ovarian tumors
● MRI
enhanced CT scans significantly
● MRI has higher sensitivity toward discovering ● MRI detecting ovarian cancer includes higher
blood-related products cost, long acquisition time, and resultant motion

artifact
5. Skin cancer as MRI MRI offers extra ordinary soft tissue contrast with Lower sensitivity that usually requires long scan-
stated by good spatial resolution (<100 m). Hence it ning period even with exogenic contrasting
Hong et al. discloses structural abnormality of skin tumors agents
[20]
Table 9.5 Mutation table
S. Name of the method Category Description Advantages

no.
1. Whole genome sequen- De novo Whole-genome sequen- The method suggests
cing as stated by muta- cing was conducted that GJC1 a DNM-
Jiang et al. [1] tion for as good as 23 implicated gene
families belonging to could possibly be a
two cohorts having probability gene for
siblings and parents Schizophrenia and its
with no impact. SCZ respective responsi-
patients were found bility could indulge
to have couple of in fetal and prema-
nonsense de novo ture neuro-
mutations (DNMs) in development, an en-
GJC1 and dangered time span
HIST1H2AD for advancing dis-
eases such as Schi-
zophrenia
2. Differential expression Mutation It inspected the RNA- The method explains
of triggering receptor in seq data in the par- the expression be-
on TREM2 cells in TRE- ietal lobe tissue of the longing to the three
irregular-AD cases M2 brain from AD pa- transcripts
versus ADTREM2 tient having TREM2 TREM2 in the post-
carriers as stated by variants from three mortem of human
Aguil et al. [42] different datasets ex- brain and also in-
pending Kallisto as cludes the resolvable
well as R package Transcript, which does
import to establish not incorporate a trans
the count of every membrane domain
transcript as well as It also observed the AD-
quantify abundant risk types inclined to
transcript in terms of the expression of ex-
transcripts per mil- plicit transcripts
lion
polymerase chain-reaction (RT-PCR) is the universal technique that is applied to

create a complete detection of SARS-CoV-2 disease. But regrettably, RT-PCR
results could be altered by inaccurate sampling and minimal virus burden.
Therefore, radiological investigations, particularly chest computed tomography
(also known as CT) scans, play an essential part in diagnosing the initial stage of
lung disease, assessing the infection advancement, and directing medical conclu-
sion for COVID-19 patients. Now owing to the minimal divergence of the disease
areas in the CT images and significant alterations in both forms and locations of
nodules in various patients, the demarcation of infected areas in CT scans of the
chest is extremely tricky. Li et al. had described primary encounters of tools that
are based on AI to detect COVID-19 pneumonia on radiological illustrations. There
are many ongoing research works and out of those we would like to put forward
few researches. Zhang et al. [27] discovered a uAI Intelligent Assistant Analysis
Table 9.6 Algorithms for hybrid biomedical data having biomedical data and
array data together
S. Name of the Category Description Advantages Limitations

no. method
1. A gradual ana- Review Compares how The main ad- Detailed work is
lysis invol- paper different pro- vantage is missing in the
ving brain cessing that Allen development of
transcrip- choices influ- Human the techniques
tome and ence the re- Brain Atlas meant for the
neuroima- sulting data of (AHBA) spatial correla-
ging as sta- brain wide works to- tions of data as
ted by transcriptomic ward a uni- well as a stan-
Aurina [43] and neuroima- fied data dardized work-
ging data processing flow is very
pipeline to much needed to
assure con- guarantee repro-
sistent and ducibility
reproducible
results
2. Statistical and Review It performs inte- The main aim Though deep
machine paper grative exami- of the ima- learning models
learning nations that ging geno- are phenomen-
methodolo- evaluate ge- mics is to ally effective in
gies for the netic changes, obtain an ad- stating data-
brain ima- molecular fea- ditional per- driven problems
ging geno- tures seized by ception into in medicine, but
mics as the several the phenoty- they were not
stated by omics data and pic traits as broadly utilized
Shen and brain imaging well as the in the brain
Thompson quantitative genetic and imaging geno-
[6] traits (QTs) molecular mics because of
that are paired mechanisms the restricted
with other bio- of our brain sample range
marker like and their ef- and elevated di-
clinical and fect on the mensionality of
environmental normal and the prevailing
data on disor- imaging and the
dered brain genomics data-
function and sets
behavior
3. An evaluation Review Provides thor- Radio- Gene expression
of radio- paper ough picture of genomics and signaling
genomics in the tumor provides ac- pathways are
oncology through radio- curate ima- extremely com-
[44] mics and ging biomar- plex
radio- kers, substi-
genomics. It tuting for
offers precise genetic test-
imaging asso- ing, also pre-
ciated with dicts risks
S. Name of the Category Description Advantages Limitations

no. method
genetic ex- and out-
pression thus comes and
serves as an also used for
alternative for personalized
the genetic treatment
testing options
4. Detecting Alz- Hybrid bio- A graph-based The graph- The main chal-
heimer’s dis- medical SSL technique based SSL lenges are to
ease using data ana- to assimilate approach characterize and
early mild lysis data of multi- with data of authenticate bio-
cognitive model modal imaging multimodal markers of AD
impairment of brain as imaging of progression that
well as to pick brain leads to an en-
the reasonable hanced early di-
predictors agnosis at early
based on ima- symptomatic
ging to opti- and or more
mize the specifically pre-
forecast preci- symptomatic
sion stages
System, software based on deep learning, to inevitably obtain and evaluate areas
believed to be affected with the virus. Then Silva et al. [45] came up with a voting-
based approach and a cross-dataset analysis called as Efficient-CovidNet. Here the
images from a given patient are categorized as a group in a voting system. The
method is examined in the two biggest datasets of COVID-19 CT analysis with a
patient-based split. To the best of our knowledge the two datasets are SARS-CoV-2
CT scan dataset and COVID-19 CT dataset. A cross-dataset analysis is also intro-
duced to evaluate the strength of the models in a more accurate scenario in which
data come from various divisions. Elaborate discussion on COVID-19 detection
patterns using CT images is given in Table 9.7.
9.7 Advent of smart technologies during COVID-19

Ever since the breakout of COVID-19, isolating and distancing have been the two
most prevalent practices that we have been made to familiarize. As a COVID-
infected patient will be isolated while the therapy still needs to continue, healthcare
turning smart was the need of the hour. The current scenario, hence, calls for AI
hand in hand with IoT to lead us to a better tomorrow, offering healthcare services
to the furthest possible patient. While we were reviewing many studies, we
observed that some of the major work has been done in IoT and AI to leverage the
Table 9.7 COVID-19 detection pattern using CT
S. Name of the method Description Advantages Disadvantages

no.
1. The uAI Intelligent As- An aggregate of 2 460 RT-PCR tests Post-CT chest the uAI Intelligent ● Movement of patients to undergo
sistant Analysis Sys- were carried on COVID patients and Assistant Analysis System CT scan is a limitation
tem the uAI Intelligent Assistant Analysis could precisely assess the ● uAI system needs manual adjust-
System evaluated the CT scans pneumonia in COVID-19 vic- ment, hence thorough investiga-
tim tion is required for assessing
probable effect
2. Efficient-CovidNet [45], Images from a given patient are cate- ● Recommends a superior yet Lack of good quality datasets, which
along with a voting- gorized as group in a voting system. efficient deep-learning model affects latency
based approach and a The method is examined in the two for the testing of COVID-19 in
cross-dataset analysis biggest datasets of COVID-19 CT CT scans
analysis with a patient-based split. A ● Reports various queries
cross-dataset analysis is also intro- regarding the two biggest
duced to evaluate the strength of the datasets
models in a more accurate scenario in ● Proposes a voting-based
which data come from various divi-
estimation method
sions
3. Active contour model Applies partial differential equation to Overall performance is good in Mainly fails when objects inside
evaluate initial contour that is applied terms of speed and accuracy images have different intensity
to 2D lung CT scan to obtain affected values that can be confronted by
portion using multistage segmentation
healthcare service. While the IoT was used mainly used for monitoring a patient
remotely, AI, on the other hand, was mainly used to break down healthcare data
and used to track and screen patients.
9.7.1 Building ML models to diagnose COVID-19

ML, being a subset of AI, has shown a lot of promise and is immensely used across
industries. In healthcare industry, it is primarily used to screen and diagnose dis-
eases. Categorically there are three main stages, the usage of ML has been broken
down into. To begin with, the first stage is Assessment followed by Tracing and
Projection and then comes Medical Aid. The Assessment is a very crucial stage
where in ML applications help a lot by analyzing the medical imaging data, like CT
scan of chest. The next stage, Tracing and Projection helps in predicting the number
of cases and tracking down the chain of contacts. To Finish, Medical Aid helps in
understanding the protein sequences and gene structure of the virus in case a
remedy could be found in tackling it either by inoculation or medicine. ML systems
can be designed such that it can emulate human intelligence. ML models have been
widely trained over datasets that comprise CT scans, MRI, and X-ray images, to
detect the irregularities.
We have come across many surveys during the review that has implemented
ML and achieved fruitful results. Amongst many studies, there happens to be one
such literature by Sandhu et al. [46] that demonstrates the way ML model uses
cloud computing and GPS technology along with Google Maps to render the
COVID-19-infected patients along the selected route and instead propose an
alternative route that has relatively less or no infected patients, thus ensuring safety
and vindicating the spread of the virus. The model, with time, has enriched itself to
reach an accuracy of around 80 percent in proposing an alternative route having
minimal infected patients. Another proposal by Choi et al. [59] did an analysis of
the sentiments of people to deduce the public reaction from social media and other
media houses. It mined the feelings from emoticons used in social media, and based
on that the ML model supervises this to suggest the actions in order to reduce
anxiety and anguish regarding the MERS-CoV.
Debnath et al. [47] proposed a way wherein ML can be used to improve the way
clinical decisions are made in the ongoing COVID-19 pandemic. Amusingly, in some
of the instances, it empowers the scientists to predict the propagation of virus across
different regions. In addition to this, ML systems are used to discover the medicine
and even at times develop inoculations, thus achieving an antidote for the virus.
9.7.2 Impact of IoT in healthcare

IoT delve into new aspects in patient care using various real-time health monitoring
and tracking, thereby allowing seamless communications among the healthcare
professionals and patients.
Taiwo et al. [60] proposed a smart healthcare support system that can be used
being at home. It monitors patients’ health and help to fetch the doctor’s pre-
scription while at home. From a doctor’s perspective, the doctor need not visit the
patient instead can use this to diagnose the disease and carry on the prognosis based
on the data accumulated remotely. Hence with this, patients can continue to be at
their home and get consulted by doctors by sharing their day-to-day health
symptoms.
The authors in [49] proposed a technique to track patients and staff and thereby
lessen the waiting period. With the usage of smart devices like blood gas analyzers,
thermometer, smart bed, glucometer, US, and X-rays, there is a drastic improve-
ment in inpatient care. It builds an integrated information system in a hospital
where all events are stowed digitally, and other data analytics are utilized toward
problem-solving throughout COVID-19 pandemic. By means of constant examin-
ing the health status, it notifies almost every forthcoming ailment and offers a
solution for its inhibition. It is beneficial in the exposure of an asthma attack and
also as a reminder of on-time medication.
9.8 Conclusion
Hence, in our book chapter, we have provided a comprehensive detailed survey
(road map) on various biomedical imaging and array-based sequence data analyses
along with their integrated study for different tissue-specific dreadful diseases
(such as cancer). We also include a brief review of ML and IoT that helped in
paradigm shift in leveraging medical facilities and incorporating smartness of
healthcare for COVID-19 treatment.
In this regard, initially, we provided different algorithms on biomedical ima-
ging stated by Bej et al. [8] that included multilevel thresholding [9,10], MRF
model [11], and clustering algorithms [12]. The advantages, shortcomings, and
usages of different algorithms of biomedical imaging are demonstrated here.
Second, we yielded different array-based sequence data analysis algorithm that is
very much useful for biomedical research. These array-based data might be divided
into different categories, viz., single-cell sequencing data, RNA-seq as stated by
Shen and Thompson [6] data, microarray data, etc. Various algorithms applied to
each category of the sequence data have been notified here. This includes various
computational problems such as gene signature discovering, biomarker discovery,
disease classification or disease subtype classification, differential expression or
differential methylation analysis, prognosis study, gene module detection [13,14],
multimodal data integration, etc. For each problem, specific algorithms (or, tools)
have been mentioned here for the benevolent of the new researchers in this domain.
The advantages, limitations, and overall usages of each category of the algorithms
along with the corresponding comparative study of the different categories of
algorithms are also described here. As a future work, we will develop a new
algorithm using ML and deep learning applied to integrated data consisting of
array- and image-based profile together. Moreover, we will also perform some
computational work on de novo mutation.
Hence, various algorithms from different backgrounds can be utilized together
to solve computational problem in big biological data. Moreover, IoT and ML
approaches are used for single RNA-seq data for COVID-19. As a future scope, we
will try to comprehensively illustrate the transcriptional changes in peripheral
blood mononuclear cells and also the risks of COVID-19 virus on different
respiratory system conditions during the recovery stage of COVID-19 by scRNA-
seq technique and scRNA data analysis.
References
[1] Jiang S., Zhou D., Wang Y., et al. ‘Identification of de novo mutations in
prenatal neurodevelopment-associated genes in schizophrenia in two Han
Chinese patient-sibling family-based cohorts’. Translational Psychiatry. 2020.
[2] Geng C., Ning B. and Shi T. ‘Single-cell RNA-seq technologies and related
computational data analysis’. Frontiers in Genetics. 2019.
[3] Kaushal C., Bhat S., Singla A., et al. ‘Recent trends in computer assisted
diagnosis (CAD) system for breast cancer diagnosis using histopathological
images’. IRBM. 2019; 40: 211–227.
[4] Mann R., Cho N. and Moy L. ‘Breast MRI: State of the art’. RSNA
Radiology. 2019.
[5] Chandrashekar V., Sureshkumar V., Kumar T. S. and Shanmugapriya S.
‘Disease prediction based on micro array classification using deep learning
techniques’. Microprocessors and Microsystems. 2020; 77.
[6] Shen L. and Thompson P. ‘Brain imaging genomics: Integrated analysis and
machine learning’. Proceedings of IEEE. 2019; 108: 125–162.
[7] Sayed E., Dahshan A., Mohsen H., et al. ‘Computer-aided diagnosis of
human brain tumor through MRI: A survey and a new algorithm’. Expert
Systems with Applications. 2014; 41: 5526–5545.
[8] Bej S., Galow A., David R., et al. ‘Automated annotation of rare-cell types
from single-cell RNA-sequencing data through synthetic oversampling’
BMC Bioinformatics. 2021; 22: 557.
[9] Bandyopadhyay S. and Mallik S. ‘Integrating multiple data sources for
combinatorial marker discovery: A study in tumorigenesis’. IEEE/ACM
Transactions on Computational Biology and Bioinformatics. 2018; 15:
673–687.
[10] Hirotaka M., Kiryu H., Furusawa C., et al. ‘SCODE: An efficient regulatory
network inference algorithm from single-cell RNA-seq during differentia-
tion’. Bioinformatics. 2017; 33: 2314–2321.
[11] Kaur R., Juneja M. and Mandal A. ‘Computer-aided diagnosis of renal
lesions in CT images: A comprehensive survey and future prospects’.
Computers & Electrical Engineering. 2019; 77: 423–434.
[12] Chaudhuri K., Sharma R. and Halarnker P. ‘Kidney and tumor segmentation
using U-Net deep learning model’. 5th International Conference on Next
Generation Computing Technologies (NGCT-2019). SSRN. 2020.
[13] Colaprico A., Olsen C., Matthew H., et al. ‘Interpreting pathways to discover
cancer driver genes with Moonlight’. Nature Communications. 2020; 11.
[14] Maulik U., Mallik S., Mukhopadhyay A. and Bandyopadhyay S. ‘Analyzing

large gene expression and methylation data profiles using StatBicRM:
Statistical biclustering-based rule mining’. PLoS One. 2015; 10: e0119448.
[15] Mallik S. and Zhao Z. ‘Detecting methylation signatures in neurodegenera-
tive disease by density-based clustering of applications with reducing noise’.
Scientific Reports. 2020; 10.
[16] Baran Y., Bercovich A., Pedros A., et al. ‘MetaCell: Analysis of single-cell
RNA-seq data using K-NN graph partitions’. Genome Biology. 2019; 20.
[17] Togashi K. ‘Ovarian cancer: The clinical role of US, CT, and MRI’.
European Radiology. 2003; 13: L87–L104.
[18] Bhuyan H. K., Chakraborty C., Shelke Y. and Pani S. K. ‘COVID-19 diag-
nosis system by deep learning approaches’. Expert Systems. 2021: e12776.
[19] Aslam B., Javed A. R., Chakraborty C., Nebhen J., Raqib S. and Rizwan M.
Computing. 2021: 1–17. 10.1007/s00779-021-01596-3.
[20] Hong H., Sun J. and Cai W. ‘Anatomical and molecular imaging of skin
cancer’. Clinical, Cosmetic and Investigational Dermatology. 2008; 1: 1–17.
[21] Afonso D. P. and Mascarenhas V. V. ‘Imaging techniques for the diagnosis
of soft tissue tumors’. Reports in Medical Imaging. 2015; 8: 63–70.
[22] Mallik S. and Zhao Z. ‘Multi-objective optimized fuzzy clustering for detecting
cell clusters from single-cell expression profiles’. Genes. 2019; 10(8): 611.
[23] Feng C., Liu S., Zhang H., et al. ‘Dimension reduction and clustering models
for single-cell RNA sequencing data: A comparative study’. International
Journal of Molecular Sciences. 2020; 21(6): 2181.
[24] Ren Q., Ma A., Zou Q. and Ma Q. ‘Clustering and classification methods for
single-cell RNA-sequencing data’. Briefings in Bioinformatics. 2019; 21(4):
1196–1208.
[25] Peng L., Tian X., Tian G., et al. ‘Single-cell RNA-seq clustering: Datasets,
models and algorithms’. RNA Biology. 2020; 17(6): 765–783.
[26] Hie B., Peters J., Nyquist S., et al. ‘Computational methods for single-cell
RNA sequencing’. Annual Review of Biomedical Data Science. 2020; 3:
339–364.
[27] Zhang H., Zhang J., Zhang H. H., et al. ‘Automated detection and quantifi-
cation of COVID-19 pneumonia: CT imaging analysis by a deep learning-
based software’. European Journal of Nuclear Medicine and Molecular
Imaging. 2020; 47: 2525–2532.
[28] Li A., Mallik S., Luo H., Jia P. and Lee D. ‘H19, a long non-coding RNA,
mediates transcription factors and target genes through interference of
microRNAs in pan-cancer’. Molecular Therapy – Nucleic Acids. 2020; 21:
180–191.
[29] Mallik S., Seth S., Bhadra T. and Zhao Z. ‘A linear regression and deep
learning approach for detecting reliable genetic alterations in cancer using
DNA methylation and gene expression data’. Genes. 2020; 11(8): 931.
[30] Bhadra T., Mallik S. and Bandyopadhyay S. ‘Identification of multi-view

gene modules using mutual information based hypograph mining’. IEEE
Transaction on Systems. 2019; 49(6): 1119–1130.
[31] Kandimalla R., Shimura T., Mallik S., et al. ‘Identification of serum miRNA
signature and establishment of a nomogram for risk stratification in patients
with pancreatic ductal adenocarcinoma’. Annals of Surgery. 2020.
[32] Qin G., Mallik S., Mitra R., Li A., Jia P. and Eischen C. M. ‘MicroRNA and
transcription factor co-regulatory networks and subtype classification of
seminoma and non-seminoma in testicular germ cell tumors’. Scientific
Reports. 2020; 10: 852.
[33] Mallik S., Odom G., Gao Z., Gomez L., Chen X. and Wang L. ‘An evalua-
tion of supervised methods for identifying differentially methylated regions
in Illumina methylation arrays’. Briefings in Bioinformatics. 2018; 20(6):
2224–2235.
[34] Bandyopadhyay S., Mallik S. and Mukhopadhyay A. ‘A survey and com-
parative study of statistical tests for identifying differential expression from
microarray data’. IEEE/ACM Transactions on Computational Biology and
Bioinformatics. 2014; 11(1): 95–115.
[35] Gomez L., Gabriel J., Young G., et al. ‘CoMethDMR: Accurate identification of
co-methylated and differentially methylated regions in epigenome-wide association
studies with continuous phenotypes’. Nucleic acid Research. 2019; 14(17): e98.
[36] Li A., Jia P., Mallik S., et al. ‘Critical microRNAs and regulatory motifs in
cleft palate identified by a conserved miRNA-TF-gene network approach in
humans and mice’. Briefings in Bioinformatics. 2020; 21: 1465–1478.
[37] Mallik S. and Bandyopadhyay S. ‘WeCoMXP: Weighted connectivity
measure integrating co-methylation, co-expression and protein-protein
interactions for gene-module detection’. IEEE/ACM Transactions on
Computational Biology and Bioinformatics. 2020; 17: 690–703.
[38] Cao B., Zhao J., Yang P., et al. ‘Multiobjective feature selection for
microarray data via distributed parallel algorithms’. Future Generation
Computer System. 2019; 100: 952–981.
[39] Shukla A. and Tripathi D. ‘Detecting biomarkers from microarray data using
distributed correlation based gene selection’. Genes and Genomics. 2020;
42: 449–465.
[40] Jia L., Peng Q., Chen X. and Sun Z. ‘A multi-objective heuristic algorithm
for gene expression microarray data classification’. Expert System with
Applications. 2006; 59: 13–19.
[41] Nikken J. and Krestin G. ‘MRI of the kidney—state of the heart’. European
Radiology. 2007; 17: 2780–2793.
[42] Aguila J., Benitez B. A., Li Z. and Dube U. ‘TREM2 brain transcript-
specific studies in AD and TREM2 mutation carriers’. Molecular
Neurodegeneration. 2019; 14.
[43] Arnatkeviciute A., Fulcher B. D. and Fornito A. ‘A practical guide to linking
brain-wide gene expression and neuroimaging data’. NeuroImage. 2019;
189: 353–367.
[44] Gullo R., Daimiel I., Morris E. and Pinker K. ‘Combining molecular and
imaging metrics in cancer: Radiogenomics’. Insights into Imaging. 2020; 11: 1.
[45] Silva P., Luz E. and Getal S. ‘COVID-19 detection in CT images with deep
learning: A voting-based scheme and cross-datasets analysis’. Informatics in
Medicine Unlocked. 2020; 20.
[46] Sandhu R., Sood S. K. and Kaur G. ‘An intelligent system for predicting and
preventing MERS-CoV infection outbreak’. The Journal of Supercomputing.
2016; 72: 3033–3056.
[47] Debnath S., Barnaby D. P., Coppa K., et al. ‘Machine learning to assist
clinical decision-making during the COVID-19 pandemic’. Bioelectronic
Medicine. 2020; 6: 1–8.
[48] Ozkaya U., Ozturk S. and Barstugan M. ‘Coronavirus (COVID-19) classi-
fication using deep features fusion and ranking technique’. In: Hassanien
AE., Dey N. and Elghamrawy S. (eds.), Big Data Analytics and Artificial
Intelligence Against COVID-19: Innovation Vision and Approach. Studies in
Big Data, vol. 78. Springer, Cham; 2020.
[49] Javaid M. and Khan I. ‘Internet of Things (IoT) enabled healthcare helps to
take the challenges of COVID-19 Pandemic’ Journal of Oral Biology and
Craniofacial Research. 2021; 11(2): 209–214.
[50] Gurcan M., Boucheron L., Can A., et al. ‘Histopathological image analysis:
A review’. IEEE Reviews in Biomedical Engineering. 2009; 2: 147–171.
[51] Mallik S. and Zhao Z. ‘Identification of gene signatures from RNA-seq data using
Pareto-optimal cluster algorithm’. BMC Systems Biology. 2018; 12(8): 126.
[52] Mallik S., Qin G., Jia P. and Zhao Z. ‘Molecular signatures identified by
integrating gene expression and methylation in non-seminoma and semi-
noma of testicular germ cell tumors’. Epigenetics. 2020; 16(2): 1–15.
[53] Akbari Y., Hassen H., Al-Maadeed S. and Zughaier S. M. ‘Covid-19 lesion
segmentation using lung CT scan images: Comparative study based on
active contour models’. Applied Sciences. 2021; 11: 8039.
model’. IEEE Access. 2020; 8: 159402–159414.
[55] Roy S. and Chakraborty C. ‘Panic buying situation during COVID-19 global
pandemic’. Journal of Information Technology Management. 2021; 13(2),
231–244.
[56] Chakraborty C. and Abougreen A. N. ‘Intelligent Internet of Things and
Transactions on Pervasive Health and Technology. 2021: 1–14.
[57] Gupta A, Chakraborty C. and Gupta B., ‘Secure transmission of EEG data
using watermarking algorithm for the detection of epileptical seizures’.
Traitement du Signal. 2021; 38(2): 473–479. https://doi.org/10.18280/ts.
380227.
[58] Shahid O., Nasajpour M., Pouriyeh S., et al. ‘Machine learning research
towards combating COVID-19: Virus detection, spread prevention, and
medical assistance’. Journal of Biomedical Informatics. 2021; 117.
[59] Choi S., Lee J., Kang M. G., Min H., Chang Y. S. and Yoon S. ‘Large-scale
Machine Learning of media outlets for understanding public reactions to
nation-wide viral infection outbreaks’. Methods. 2017; 129: 50–59.
[60] Taiwo O. and Ezugwu A. E. ‘Smart healthcare support for remote patient
monitoring during COVID-19 quarantine’. Informatics in Medicine
Unlocked. 2020; 20.
[61] Zhang Y., Ma Y., Huang Y., et al. ‘Benchmarking algorithms for pathway
activity transformation of single-cell RNA-seq data’. Computational and
Structural Biotechnology Journal. 2020; 18: 2953–2961.
[62] Dash S., Chakraborty C., Giri S. K., Pani S. K. and Frnda J. ‘BIFM: Big-data
driven intelligent forecasting model for COVID-19’. IEEE Access. 2021; 9:
97505–97517. 10.1109/ACCESS.2021.3094658.
[63] Baran Y., Bercovich A., Sebe-Pedros A., et al. ‘MetaCell: analysis of single-
cell RNA-seq data using K-nn graph partitions’. Genome Biology. 2019; 20:
206.
Chapter 10
Review of medical imaging with machine
learning and deep learning-based approaches for
COVID-19
Swapnil Singh1, Vidhi Vazirani1 and Deepa Krishnan1
Abstract
COVID-19 came with unprecedented challenges that disrupted activities of gov-

ernment, economies, and societies all around the world. It is caused due to severe
acute respiratory syndrome coronavirus 2. We are witnessing how governments and
administrations have taken proactive steps to limit the spread of disease. Even
though many countries have enforced strict lockdown to curtail its transmission,
mortality rates continue to rise. An early detection of COVID-19 is inevitable as it
helps in seeking early medical intervention and to minimize community spread.
COVID-19 can be diagnosed with the help of gold standard reverse transcription
polymerase chain reaction (RT-PCR) tests, quick antigen and antibody tests sup-
plemented with computerized tomography (CT) scan and X-ray imaging.
Considering the increase in the number of cases worldwide, there is a need for
economical and viable ways to detect COVID-19. RT-PCR tests continue to be
dominant; however, there are few concerns related to the sensitivity associated with
the test results. Thus, medical community underlines the importance of medical
imaging tools like CT scans and chest X-rays (CXRs) and the abnormalities
revealed in CXRs and CT scan can help in detecting COVID-19.
Medical imaging and analytic tools that use machine learning and deep
learning algorithms can enhance the diagnosis and prediction of COVID-19. There
are many effective and proven image recognition techniques in deep learning like
convolutional neural networks and transfer learning that can be used to design
promising applications for COVID-19 detection. These models can be enhanced
using image segmentation and edge detection techniques. In the proposed book
chapter, we have reviewed the impact of the COVID-19 pandemic on the global
community, the need for reliable, quick, and economical ways to detect it. We have
surveyed the existing mechanisms for COVID-19 detection using machine learning
1
Computer Engineering Department, Mukesh Patel School of Technology, Management, and
Engineering, NMIMS University, Mumbai, India
and deep learning algorithms and presented a critical review of the major short-
comings in existing mechanism that could open further research in this area.
Keywords: Machine learning; COVID-19; Deep learning; Medical imaging;

Classification
10.1 Introduction
Coronavirus affects distinct species causing varied symptoms and diseases. The
latest mutant of the coronavirus family is the novel COVID-19 virus. This mutant
has affected millions of people all over the globe. The initial cases of this virus
were observed in Wuhan, China in early December 2019. After declaring the virus
that had already affected thousands a ‘Public Health Emergency Outbreak’ on 30
January 2020, the World Health Organization declared the severe acute respiratory
syndrome coronavirus 2 a pandemic on 11 March 2020.
With the increase in spread from one country to another, COVID-19 virus has
started to show mutations that are more dangerous and hence a cause of concern.
These mutations have started to show symptoms which are not displayed by the
traditional virus. These symptoms are low-grade fever, dry mouth, non-salivation,
gastrointestinal symptoms, and headaches [1]. It is important to detect the virus as it
can be easily transmitted from one person to another. Scientists anticipate that the
virus can be transmitted via droplets or aerosols released by an infected person. The
spread is typically within a 1-m radius [2]. However, some studies suggest that
COVID-19 is airborne and can spread to greater distances. This is a cause of great
concern. Greenhalgh et al. [3] presented 10-fold reasons to support the transmission
of COVID-19 being airborne. A reason stated to prove this was the transmission of
the virus between people in different rooms. Considering the high transmission rate
of the virus, governments all over the world have applied social-distancing restric-
tions to stop the spread of the virus. But treating a person is difficult considering the
social distancing norms. So Rahman et al. [4] developed the EdgeSDN-I4COVID,
this is a human-less IoT-enabled environment working without interruption. This
architecture assures efficient and intelligent management of COVID-19.
Considering the high growth rate of COVID-19, governments were not able to
prepare for the pandemic. Dash et al. [5] used the Facebook Prophet model to
predict the value of cases for the next 90 days, including the peak. They used data
from the USA, Brazil, India, France, Russia, and the United Kingdom shows that
these are the six most worst hit countries. The model was also able to identify five
significant change points and gave a goodness-of-fit as 85% mean absolute per-
centage error (MAPE).
COVID-19 is curable if it is detected at an early stage and the appropriate
medication is given to the patient. Several ways to detect COVID-19 are reverse
transcription polymerase chain reaction (RT-PCR), rapid antibody testing (RAT),
and imaging techniques like computerized tomography (CT) scan, ultrasound, and
Review of medical imaging 263
chest X-ray (CXR). RAT is the fastest way of detecting COVID-19, and its kits
provide results in a few minutes, but it has a lower efficiency than RT-PCR. RAT
tests are done by pricking one’s finger and putting the blood in the test kit.
However, the problem with RAT is that since it tests antibodies and thus gives
negative results when the patient is recently infected, indicating a false negative.
Another problem is that in case the person has recovered, the person can still have
antibodies present in his/her blood, which would give a positive result, indicating a
false positive. Scohy et al. [6] got an efficiency of 30.2% for RAT tests. Thus, they
concluded that even though RAT tests give fast results, they are not reliable.
RT-PCR tests are more accurate than RAT tests, as proved by authors of [6],
but they are costly and take a lot of time. Considering a highly populated country
like India, RT-PCR test results take days to reach the patient; this creates life-
threatening concerns for the patients and people around them. Another concern
with RT-PCR test is that it is not able to detect the new mutants of COVID-19.
Thus, there is a need to explore other options which are reliable and more efficient.
This is where radiology comes into the picture. Pneumonia being the most common
symptom of COVID-19 can be detected with the help of CT scans and CXRs.
Chendrasekhar et al. [7] performed a comparative study on the efficiency of chest
CT scans and RT-PCR test. They performed their study on 1,400 individuals out of
which 41.6% were males and the average age was 46.282.7 years. They repre-
sented the results in an odds ratio (OR) with the confidence interval being 95%.
They proved that chest CT scan is superior to RT-PCR (OR 3.86, 95% CI), and it is
a rapid and more efficient alternative to RT-PCR tests.
Another radiological way of detecting COVID-19 is with the aid of a CXR.
CXRs are commonly used to detect pneumonia. CXRs can be obtained easily, as
they are cheap, readily available, and give results fast. Yasin and Gouda [8] per-
formed a study on 350 patients who were COVID-19 positive already. Out of these
350 patients, 220 showed baseline abnormalities in the CXRs. Out of the remaining
130 patients, 48 patients showed abnormalities in a follow-up. They concluded that
their findings were good for accessing COVID-19.
COVID-19 is spreading and mutating at high pace. Many counties have faced
second and third waves of COVID-19. In a country like India which has seen more
than 4 lakh positive cases per day, there is a need for an efficient, fast, and cheap
way of detecting COVID-19 so that the infected people can be treated on time. RT-
PCR tests take a lot of time for giving results and require a lot of workforce and
logistics, so it is not an optimal choice, whereas in the case of RAT tests, they have
a low efficiency rate. There comes in the field of radiography. As seen earlier, CT
scans and CXRs have established their efficiency for detecting COVID-19. CT
scans and CXRs provide a fast and efficient way of detecting COVID-19.
Machine learning and deep learning provide us with an opportunity to reduce
the burden of doctors and medical professionals. These models when trained
properly can give us accurate and efficient detection mechanisms that can help us
in an early detection of COVID-19. Training algorithms with the help of CT scans,
CXRs, and ultrasound can help us in automating the scanning process for detecting
COVID-19. Deep learning algorithms can be used to extract the important features
from these radiography images and then be used to train machine learning models.
Other feature extraction algorithms can be used to extract features, and then these
features can be used to train machine learning models. Image-processing techni-
ques can be used to enhance the radiography images before passing them through
the model for training them. These techniques can also be used to localize and
pinpoint the infected regions in CT scans, CXRs, and ultrasound images. These
applications would not only help in better diagnosis but also in reducing the burden
of medical staff, when there is a surge in the number of positive cases.
The major contributions of this survey include the following:
1. Reviewing the machine learning and deep learning architectures that are used
by researchers for the detection of COVID-19 giving emphasis to datasets used
and contributions by authors.
2. Presented a comparative analysis of the research works reviewed with impor-
tance to the challenges addressed by researchers in each work.
3. Identified the research gaps and provided future directions for the same.
This chapter is structured as follows: Section 10.1 consists of the introduction.
Section 10.2 talks about the reviewed work. The preliminary discussion gives a
brief idea about the various models and techniques discussed in the reviewed work.
Further in Section 10.3, a comparative study is done on the work as reviewed in the
previous section. Section 10.4 describes the most common research gaps;
Section 10.5 as “Conclusion” states the future scope of expanding this study.
10.2 Literature review
10.2.1 Reviewed work

As seen in Figure 10.1, VGG networks were used in maximum research papers
reviewed in this book chapter. ResNet, machine learning algorithms, and DenseNet
were also frequently used. This chapter also reviews paper which use state-of-the-
art algorithms like StackNet, Auxiliary Classifier Generative Adversarial Network,
Generalized Regression Neural Network, and Probabilistic Neural Network (PNN)
on various datasets and with a different combination of other algorithms and pro-
vide promising results.
10.2.1.1 Convolutional neural networks

Convolutional neural networks (CNNs) are in forefront of medical image analysis
and have given promising results for disease identification. There are significant
studies done using CNN on COVID-19 CXRs and CT scans. In research work [9],
authors have proposed a 22-layer CNN model which classifies X-ray images into
two class, three class, and four class. The authors created a dataset of 7,390 images.
The 22-layer CNN model was tested extensively to select an appropriate number of
layers, activation functions, and optimizers. The proposed work has been able to
achieve an accuracy of 99.1% for two-class classification, 94.2% for three-class
classification, and 91.2% for four-class classification. In comparison with many
Number of research papers per algorithm
14
14
12
11
10 9
8
Count
7 7 7
6
5
4
4
3 3 3
2
2
1 1 1 1
0
Custom CNN
VGG
Xception
ACGAN
GRNN
PNN
StackNet
GoogLeNet
ResNet
DenseNet
SqueezeNet
AlexNet
InceptionV3
InceptionResNetv2
NasNet
ML algorithms
MobileNetV2
Algorithms
Figure 10.1 Number of research papers per algorithm
related works the authors’ research works have attained better score in accuracy,
precision, recall, F1 score, sensitivity, and specificity. However, the model per-
formed poorly in low-quality X-ray images.
Another research work [10] in the same line is by authors Hernandez et al.
where they have used CNN with non-COVID-19 pulmonary X-rays and tuning
the final layers with COVID-19 X-ray images. The authors have accounted for the
data scarcity problem for COVID-19 images through augmentation techniques
available in OpenCV. The first approach used by the authors was to train CNN
model using pulmonary X-rays from CheXpert. Later final layers were added and
trained using CNN with COVID-19 X-ray images. However, the accuracy, pre-
cision, recall, and F1 score are found to be 0.68, 0.67, 0.63, and 0.65, respec-
tively. The second approach used by authors is with transfer learning using
ResNet50, VGG16, and DenseNet121. ResNet50 architecture is found to have
better accuracy, precision, recall, and F1 score of around 90%. The authors have
also extended the work as a multi-classification task with three classes of normal,
COVID-19, and pneumonia. The testing accuracy for VGG16 is found to be
above 0.91 for the three-class classification scenarios and is less prone to over-
fitting than the other models. They also extended the work with DenseNet121 and
are found to have more false predictions.
In [11] also, a custom CNN and transfer learning using ImageNet models are
trained on publicly available datasets. In customized CNN approach, the authors have
reduced the model parameters using the separable convolutions. They have also used
the Talos optimization package to optimize the parameters and hyper parameters.
The iterative model pruning technique is used to find the optimal number of neurons
in the convolutional layers. The custom model gave an accuracy of 0.9467 and an
AUC score of 0.9842. It is found that VGG16 and VGG19 based models are found to
have the highest accuracy and AUC score. When the transfer models are applied on
binary classification task of detection between normal and COVID-19, it gave very
high-performance measures. In the three-class classification task for normal, bac-
terial pneumonia or COVID-19 pneumonia InceptionV3 gave an accuracy of 0.9742
and an AUC score of 0.9969. One of the significant contributions made in this pro-
posed work is iterative pruning of the specific models which helps in reducing the
number of trainable parameters. They have also improved the performance by using
ensemble of models. The performance of the model is further validated by using
visualization techniques that verify the localization of the ROI of the input dataset.
10.2.1.2 Transfer learning

StackNet
StackNet can be used for multiple applications using the same neural network; one
such application was presented in [12]. The authors proposed a new classification
algorithm named CovStacknet. This network is based on StackNets and used
VGG16 for feature extraction. After extracting features from CXR images, they
used feature engineering to choose the key features for classification. They used a
combination of two feature selection methods, the first one was selecting only those
features whose variance was greater than a set threshold, the second one was
selecting best features using random forest algorithm. Considering the imbalance in
all the three datasets considered for this study, they used SMOTE. After these steps,
the flatten layer of VGG16 is used to convert the two-dimensional tensors to a
vector that could be fed to StackNet. The authors decided to stack logistic regres-
sion, extra trees (ET) classifier, support vector machine (SVM), linear discriminant
analysis, RF, K-nearest neighbours (KNN), bagging classifier, decision tree (DT)
classifier, adaptive boosting classifier, Gaussian naı̈ve Bayes (GNB), and gradient
boosting classifier. They used five estimators in the first layer, three in the next one,
and one in the third layer. For the first dataset, labeled OCT (optical coherence
tomography), and CXR images for classification, they classified the CXR in two
classes, normal and pneumonia, achieving an accuracy of 97%. For the second
dataset, COVID-19 CXR dataset consisting of non-COVID-19 X-rays and COVID-
19 X-rays, achieving an accuracy of 74%, and for the third dataset, COVID-19
CXR had non-COVID-19 X-rays and COVID-19 X-rays giving an accuracy of
98%. The proposed model was made to first classify normal and pneumonia;
thereafter the pneumonia X-ray was classified into COVID-19 and non-COVID-19.
GoogLeNet
GoogLeNet is a state of art algorithm developed by Team GoogLeNet. Mayya et al.
[13] explained an application of GoogLeNet along with RF to detect COVID-19.
Their work consisted of a fusion of image and text classification. For selecting the
best combination of features, they used RF and trees on the two cases, first case
with two classes which were COVID-19 and non-COVID-19, and the second case
with four classes which were COVID-19, H1N1, pneumonia, and normal. Along
with RF, they also used RUSboost Trees, Boosted Trees, Medium Trees, Bagged
Trees, Coarse Trees, and Fine Trees. The text dataset had 32 attributes, when all of
them were taken into consideration, and the accuracy for four-class classification
was 77.5% and for two-class classification was 85.8%. Next, they reduced to 28
attributes and achieved an accuracy of 84.2% for four-class classification and
90.8% for two-class classification; finally, they reduced the attribute count to 22
attributes, hence, achieved an accuracy of 83.3% for four-class classification and
two-class classification. For the image dataset, it was preprocessed by using a
median filter that was used to smooth the intensity of the images, and they used
Otsu automatic thresholding to get the target regions and clear borders and hole
filling. For image classification, they replaced the fully connected layers for two-
and four-class classifications. For the image classification model, the four-class
classification gave an accuracy of 95.5% for COVID-19 and 72.43% for other
classes, whereas in the case of two-class classification, the COVID-19 accuracy
was 80.3% and 75% for non-COVID-19 class. Similarly, for the fusion model, the
accuracy of four-class classification is 91.3%, and 91.67% for two-class classifi-
cation. Since the model is trained on both textual and image data, missing values in
the textual data should not affect the performance of the model.
VGG
VGG networks are prominent neural networks that are used in transfer learning.
One such work was presented by Panwar et al. [14], where they used
VGG16 model to produce a new model, named nCoVNet. The images in the
dataset were first resized to 224 224 3. These images were then augmented by
rotating the images in the rage of 20 degrees, flipping images horizontally and
vertically. The model consisted of an input layer, followed by 18 combinations of
convolutional layers with ReLU as activation function and MaxPooling. These
layers were part of the VGG16 model. These layers were followed by a flatten
layer, an AveragePooling2D layer, a fully connected layer with 64 neurons and
ReLU as activation function, a dropout layer with the threshold of 0.5, and then we
have an output layer with two neurons. The model was compiled using the Adam
optimizer for 80 epochs and 0.0001 as learning rate. After training they got true
positives as 41, false negatives as 1, false positives as 9, and true negatives as 33.
The training accuracy was in the range of 93%–97% and with the training loss of
0.2%. They tried to solve the problem of data leakage by splitting the dataset using
a logical method instead of the traditional train_test_split.
Another such work was presented in [15], they compared VGG16, VGG19,
ResNet50V2, InceptionV3, Xception, InceptionResNetV2, NasNetLarge, and
DenseNet121; after comparing these CNN models, the best results were found in the
case of VGG19. The fine-tuned VGG19 model was trained and tested using X-ray
images, CT-scans, and ultrasounds. Since the dataset was small, data augmentation
was performed by flipping images horizontally and vertically. To reduce the effect of
sampling bias authors decided to apply N-CLAHE, this method not only normalizes
the images but also enhances the minute details in the images. After studying the
trends in tuned hyperparameters, five experiments were conducted, each having a
separate set of hyperparameters and dataset. The output classes for each of the five
experiments were different as well. The highest accuracy observed, using the
Ultrasound dataset, was 100%. The parameters were as follows: learning rate was
105, dropout rate kept at 0.2, batch size was 2, hidden layer states were set to 64 and
100 epochs. The best accuracy is for the X-ray dataset was 86% and was achieved by
tuning the learning rate to 103 dropout rate at 0.2, batch size and hidden layer state
to 8 and 100 epochs. For CT scans, however, the epochs had been set at 70 to avoid
overfitting and batch size was reduced to 4 while hidden states were increased to 64.
After tuning the learning rate at 103 and dropout rate at 0.2, an accuracy of 84% was
achieved.
Vaid et al. [16] presented the work on VGG19 with the main aim being to
reduce the false-negative cases as they are a major cause of concern. The model
proposed in the study was a modified version of VGG19. A trainable multi-layer
perceptron (MLP) was added to the traditional model to increase accuracy. This
MLP, consisted of four fully connected layers consisting of 512, 512, 64, and 2
neurons, respectively, the output layer used SoftMax as the activation function.
After the architecture was finalized, the hyperparameters were tuned: learning rate
was set to 0.001, epochs were 100, and batch size was made 15. Dimensions of the
input image was 512 512 and the output was binary classification (COVID-19 vs
no COVID-19). The modified VGG19 model gave superior results with an accu-
racy of 96.3% and 0.151 binary cross entropy loss. The testing data gave 74 true
positives, 32 true negatives, 3 false negative, and 1 false positive. However, the
study did have some limitations. The number of samples in the dataset was small
due to the non-availability of appropriate data. Another limitation stated was that
the results were purely based on the X-ray images and did not consider the various
external factors that may influence the result.
Brunese et al. [17] presented the use of VGG16 with two models to classify the
CXRs into healthy and disease classes and other disease classes and COVID-19.
Authors proposed a three-step method: first detecting pneumonia from the dataset,
second segregating COVID-19 from pneumonia, and third localizing the area
where the infection was present in COVID-19 CXRs. They used VGG16 for
detecting pneumonia from the dataset and then detecting COVID-19 from the
pneumonia samples. They replaced the fully connected layers of VGG16 with
average pooling layer, flatten layer, a fully connected layer with 512 neurons, a
dropout layer with threshold as 0.5, another fully connected layer with 64 neurons,
and finally an output layer with 2 neurons and activation function as SoftMax.
After classification, they used Grad-CAM, where a heat map is generated using the
gradient information in the last convolutional layer, and then a heat map is created.
This heat map when added with the CXR gives the infected regions. The accuracy
for classifying healthy and disease classes was 96% and the time taken by the
model was 2.569 s, whereas the model gave an accuracy of 98% for the classifi-
cation of COVID-19 and other forms of pneumonia and the time taken by the
model was 2.498 s. For the future work, the authors aim to use this model over a
wider range of pulmonary diseases and find ways to verify the accuracy of the
model. A limitation of this chapter was that the dataset was highly biased and not
equally distributed; also the authors only restricted their work to VGG16 network
and did not explore other transfer learning models.
ResNet
ResNet CNN architectures are based on residual networks. ResNet18, ResNet34,
ResNet50, and ResNet101 are popular ResNet architectures. Minaee et al. [18]
compared ResNet18, ResNet50, SqueezNet, and DenseNet-121. For this purpose,
the authors built a custom dataset, COVID-CXR dataset, with CXR of COVID-19
and normal patients. The training set initially had just 84 COVID-19 X-rays and
had to be extended to around 420 images using data augmentation. The testing set
had 100 COVID-19 and 3,000 normal images. Even after data augmentation, the
number of training samples is limited and thus the authors only fine-tune the last
layer of each model after performing transfer learning. Each model runs for 100
epochs, has a learning rate of 0.0001, batch size is 20 and uses Adam Optimizer and
Cross-Entropy as its loss function. The loss function is minimized using stochastic
gradient descent. After training was completed, the models were tested and eval-
uated using performance metrics like sensitivity and specificity, precision–recall
curve, and confusion matrix. The best models were ResNet18 and SqueezeNet.
While all models had a sensitivity of approximately 98%, ResNet18 had a speci-
ficity of 90.7% while SqueezeNet had a specificity of 92.9%. Both ResNet18 and
SqueezeNet had only 2 false negatives, but 321 and 237 false positives, respec-
tively. The authors suggest that the model needs to be trained on larger COVID-19
datasets in order to achieve the desired performance.
More CNN models were compared by Nayak et al. [19] who concluded that
ResNet50 worked best when performing a binary classification on COVID-19
images. The authors started their study by collecting the right dataset. Images were
normalized by dividing the pixels with 255. Data augmentation was performed to
increase the number of images. Images were rotated by an angle of 5 degrees
(clockwise), horizontally flipped, scaled by 15%, and Gaussian noise (mean ¼ 0
and variance ¼ 0.25) was added to the images. The eight models compared by the
study were InceptionV3, MobileNet, AlexNet, GoogLeNet, VGG16, SqueezNet,
ResNet34, and ResNet50. Images were resized to 224 224 for all models expect
for InceptionV3 which has an input size of 299 299. Apart from different models,
different hyperparameters (optimizer, learning rate, epochs, and batch size) were
tested on each model to understand what worked best. For optimizers, Adam,
RMSProp, AdaDelta, and SGD were tested. Adam worked best for most models. A
batch size of 32 gave the best results for all models expect VGG16 and MobileNet
and thus was selected for the study. After comparing all the models using different
performance metrics, it was found that ResNet models gave 100% sensitivity.
However, ResNet34 outperformed ResNet50 due to its higher accuracy. While
ResNet34 gave an accuracy of 98.33%, ResNet50 stood at 97.50%.
Authors of [20] presented a combination of ResNet50 and U-Net for detecting
COVID-19 from CT scans, naming the network CoVNet. The ResNet50 model was
modified and after all the convolution layers, the final feature maps were fed to a
fully connected layer, giving probability for each class. The dataset contained only
those CT scan slices of thickness less than 3 mm. From the CT scan images, with
the help of U-Net, the lung area, which was the region of interest, was extracted and
used for training the model. For COVID-19 class, the sensitivity, specificity, and
AUC score were 90%, 96%, and 0.96, respectively, whereas the pneumonia class
gave a sensitivity of 87%, specificity of 92%, and the AUC score was 0.95.
Similarly, the AUC score for normal class was 0.98, specificity was 96%, and
sensitivity was found to be 94%. For better interpretation, the authors used Grad-
CAM for highlighting the infected regions in the CT scans. Some limitations stated
by the authors about their studies are as follows: (i) since COVID-19 shows
symptoms similar to that of pneumonia; (ii) the lugs could be affected by multiple
infections at the same time, so it is not possible to differentiate these images just
based on the visual appearance; (iii) a large dataset can be used for the study, for
creating a better and generalized model; (iv) deep learning models are not
interpretable and not transparent.
Ismael et al. [21] compared three ways to detect COVID-19 cases using CXRs:
deep feature extraction followed by SVM classification, fine-tuning pre-trained CNN
models, and building a custom end-to-end trained CNN model. In the first method,
the various kernels used in combination with SVM were linear (epsilon ¼ 0.04),
quadratic (epsilon ¼ 0.02), cubic (epsilon ¼ 0.01), and Gaussian (epsilon ¼ 0.01).
VGG16, VGG19, ResNet18, and ResNet101 were also used for feature extraction;
however, the results were not satisfactory enough. ResNet50 gave an average accu-
racy of 92.6% (all kernels combined). The ResNet50, combined with the linear
kernel SVM, gave an accuracy of 94.7%. In the case of fine-tuning, the same CNN
models were tuned and tested. Again, ResNet50 gave the highest accuracy of
94.63%, two false negatives, and five false positives. The custom CNN gave an
accuracy of 91.83% which was not as good as the previous models. Thus, the best
results were obtained using ResNet50 in combination with linear kernel function
of SVM.
In a study proposed by Zhong et al. [22], ResNet50 was used as base model for
feature extraction from CXR images. The authors began by collecting data from
three sources: COVIDx, five hospitals from Partners HealthCare system in the
United States, and four hospitals in South Korea. There were three classes in this
study: COVID-19, non-COVID-19 pneumonia, and control (normal images). After
performing essential preprocessing steps, segmentation was performed using an
ensemble of five deep neural networks based on the EfficientNet architecture.
ResNet50 is used as the backbone of the main model. Features are extracted in two
parts of the model. The first part, F1 consists of the first 22 layers of the ResNet50
architecture whereas the second part, F2 contains the next 18 layers. Between the
two parts, there is a special attention module placed that generates a 16 16 mask
applied on the output of the ResNet50 model. After the two parts and attention
model, the features are mapped and passed through the remainder layers of the
ResNet50 architecture. To train this model, the Adam optimizer was used along
with a learning rate of 3e5, batch size of 48, 5 epochs, and 2,000 iterations. The
aim of the entire model is to learn function f that embeds given image into a
d-dimensional space. The next step is to classify the images using the features. To
do so, KNN algorithm is used with k ¼ 10. The KNN is found using the cosine-
similarity measure:

Si;j ¼ h f ðxi Þ; f xj i=kf ðxi Þk2 kf xj k2 (10.1)
However, to normalize the embedding, the denominator is set to 1. To calcu-
late the loss in the final paired metric learning stage, the multi-similarity loss
function in the following equation is used:
" #
1X m
1 X
aðSi;j lÞ
L ¼ log 1 þ e
m i¼1 a
" j2Pi
# (10.2)
1 X
bðSi;j lÞ
þ log 1 þ e
b j2Ni
where xi is the anchor image, Pi is the images with same label, Ni is the images with
different label, m is the batch size, a, b, l are the hyperparameters.
To better understand the effectiveness of the proposed architecture, the results
were compared with the baseline ResNet50 model. While the proposed model had
an accuracy of 83.9%, ResNet50 showed an accuracy of 81.5% on the same data.
The proposed model was also more sensitive towards the COVID-19 inputs.
DenseNet
DenseNet or Dense Convolutional Network connects one layer to all the layers in
the sequence ahead. Authors of [23] proposed a comparative study wherein CNN
models were used for extracting features followed by classification done by
machine learning techniques. For the study, the authors custom-made two datasets
– Dataset A and Dataset B. The only difference between the two datasets was that
they had normal X-ray images from different sources. To increase the size of
training samples, data augmentation using the affine transformations was per-
formed. A total of 144 experiments were conducted under this study. Each had a
different pair of CNN model and classifier. Since both datasets had different
samples, different models worked best for them. For Dataset A, DenseNet121 with
MLP worked best. But for Dataset B, DenseNet201 along with MLP worked.
Accuracy scores for the two datasets were 98.974% (approx.) and 95.641%
(approx.), respectively. However, when data was not augmented, Dataset A gave
best results with MobileNet and SVM (linear) and gave the highest accuracy. They
took into consideration many performance metrics like confusion matrix, F1 score,
false-positive rate, and time taken by network for training.
In [24], the authors have proposed a three-step model to detect COVID-19

using CT scan. The process starts with data augmentation. Images were rotated and
flipped. The second step is to perform deep feature extraction using the
DenseNet201 model as it gave a higher accuracy as compared to other CNN
models. The third step is to perform classification. For this purpose, the traditional
machine learning models did not give satisfactory results. However, the multi-
kernels extreme learning machine-based deep neural networks (MK-ELM-DNN)
gave impressive results. The dataset used had two classes (non-COVID-19 and
COVID-19). The DenseNet201 model for the current study contains 709 layers,
806 connections, 201 depths, and about 20 million parameters. The extreme
learning model (ELM) is a neural network with one hidden layer. It learns at a
faster rate and performs better generalization as compared to traditional neural
networks. After comparing various functions, the ReLU function was selected
alongside ELM model. An accuracy of 98.36% was achieved using the MK-ELM-
DNN model. Other metrics were – sensitivity: 98.28%; specificity: 98.44%; pre-
cision: 98.22%; F1-score: 98.25%; and AUC: 98.36%.
Ohata et al. [25] proposed Casecade-SEMEnet which was developed using
SEME-ResNet50 and SEME-DenseNet169. There were three datasets used in this
study for three different purposes. Dataset 1 was divided into three classes – bac-
teria, virus, and normal. The ResNet50 model was trained using this dataset and its
goal was to categorize the X-ray into one of the three classes. Once it was detected
that the X-ray belonged to the virus class, it was passed on to the next CNN model –
DenseNet169. To train this model, Dataset 2 was used which had two classes –
COVID virus and other viruses. The third dataset, U-Net was used to remove any
non-pathological features on the X-ray films. This helped a better classification of
X-rays from unknown sources as it performed segmentation. The biggest challenge
that the authors faced here was that the size of the image in the ImageNet database
which was used to train the CNN models was very small as compared to the
training X-ray images. Thus, global average pooling was applied to all input ima-
ges. SE-Structure was added, and Attention module was used to channel the char-
acteristics. These additions boosted the accuracy. To further enhance results,
CLAHE was added to the training set and MoEx structure was used in networks.
Dataset 1 gave an accuracy of 85.62% with SEME-ResNet50, and Dataset 2 gave
an accuracy of 97.14% on the SE-MoEx-DenseNet169.
SqueezeNet
SqueezeNet provides similar efficiency as AlexNet. Chowdhury et al. [26] presented
the comparison between SqueezeNet, AlexNet, DenseNet201, and ResNet18 for the
detection of COVID-19 with the help of CXRs. These X-rays were collected from
various sources and then resized to 224 224 for DenseNet201 and ResNet18 and
227 227 for AlexNet and SqueezeNet. Since the number of images in the dataset
was low, the images were augmented and the normalized. Mini batch gradient decent
was chosen as the optimizer, the learning rate was set to 0.0003, momentum as 0.9,
and 16 mini-batches. For two-class classification, ResNet18 gave the testing accu-
racy of 96.7%, AlexNet gave the testing accuracy of 97.5%, whereas the testing
accuracy of DenseNet201 was 98%, and the highest testing accuracy was 98.3%.
SqueezeNet gave the AUC as 0.998 and F1 score as 0.983. For three-class classifi-
cation, the testing accuracy for ResNet18 was 95%, testing accuracy of AlexNet was
95.4%, whereas DenseNet201 gave the testing accuracy as 96.7%, and again the
highest testing accuracy was given by SqueezeNet as 98.3%. The AUC score was
0.99 and F1 score was 0.983 for SqueezeNet. The performance of this chapter could
be improved by using more images in the dataset, which would help to generalize the
models.
Xception
Xception architecture is based on the Inception CNN architecture and is short for
‘Extreme Inception’. Khan et al. [27] presented the use of Xception architecture to
build a model – CoroNet – to detect COVID-19 using CXR images. In addition to
the traditional Xception architecture, CoroNet has a dropout layer and two fully
connected layers at the end of the model. After finalizing the architecture, three
models were prepared. The first and most elaborate set was a four-class classifi-
cation (COVID-19, Viral Pneumonia, Bacterial Pneumonia, and Normal). The
accuracy of this classification was 89.6%, F-measure was 89.8%, precision of 90%,
recall 89.92%, and specificity 96.4%. The other two models were based on this
main multi-class classification model. They were used to perform three-class
(COVID-19, normal, pneumonia) and two-class binary (COVID-19, normal) clas-
sifications. While the accuracy of the three-class model was 95%, the model made
for binary classification gave an accuracy of 99%. Although CoroNet provides a
way to detect the deadly virus in a less expensive manner and gives good accuracy
especially with binary classification, it needs to be trained and tested using bigger
datasets before it can be used practically.
Jain et al. [28] after comparing InceptionV3, ResNeXt, and Xception found
that the best results were obtained using the Xception model. The dataset used in
the comparison was gathered from Kaggle. The model proposed was for a three-
class classification (COVID-19, Normal, and Pneumonia). Preprocessing was done
using the Keras Data generator. The data was then used to train the model. The
CNN models used LeakyReLU for the activation function and SoftMax for the final
classification. While the Xception model gave an accuracy of 100% for the training
data, the testing data had an accuracy of 97%. The other two models, Inception and
ResNet, had a testing accuracy of 96% and 93%, respectively. The biggest limita-
tion of this study, however, was that the dataset was highly imbalanced.
In a more elaborate study as proposed by El Gannour et al. [29], six dif-
ferent CNN models were compared – VGG16, VGG19, InceptionV3, Xception,
ResNet50V2, and MobileNetV2. The ‘COVID-19 Radiography’ dataset used for
this study was curated by University of Doha and is publicly available on
Kaggle. Transfer learning was used to train the model on the COVID-19 dataset.
After preprocessing, the images were fed to these models for training.
Hyperparameter tuning was performed: learning rate ¼ 0.0001, batch size ¼ 15,
and epochs ¼ 50. While most models gave an accuracy of 97%, InceptionV3,
and Xception gave an accuracy of 98%. However, Xception proved to be the
better model as it had 100% COVID-19 sensitivity compared to the 95% sen-
sitivity displayed by the inception architecture.
MobileNetV2
MobileNetV2 is an improved version of MobileNetV1 and works on the concept of
ResNet. Apostolopoulos et al. [30] presented the use of MobileNetV2 and other
transfer learning algorithms to compare their efficiency in detecting COVID-19
from CXRs. For this purpose, they used two datasets, containing three classes
which were normal, pneumonia, and COVID-19. For their study, they used
VGG19, MobileNetV2, Inception, Xception, and InceptionResNetV2, and they
compared the performance of these models by using ReLU activation function for
all layers, a dropout layer between two fully connected layers, Adam optimizer to
compile the model, and trained the model for 10 epochs with the batch size 64.
VGG19 gave an accuracy of 93.80% on the first dataset whereas the accuracy for
MobileNetV2 was 92.85% on the same dataset. Similarly, on the same dataset,
Inception, Xception, and InceptionResNetV2 gave an accuracy of 92.85%. The two
best performing algorithms, VGG19 and MobileNetV2, were applied on the second
dataset. Here MobileNetV2 outperformed VGG19 and gave an accuracy of
94.72%, sensitivity of 98.66%, and specificity of 96.46%. MobileNetV2 is better
than VGG19 because it gives a better specificity, thus proving to better for the
classification task on a particular dataset. A limitation to the study was that all five
algorithms were trained for only 10 epochs, training the models further could
increase the efficiency.
An excellent example of MobileNetV2 being deployed in a mobile application
was demonstrated by Li et al. [31]. Authors used deep learning to build an applica-
tion that would detect COVID-19 from CXRs. They use a two-stage detection
mechanism where DenseNet121 is used as a pretrained attending physician and
resident doctor and MobileNetV2 and SqueezeNet are used as medical students. The
goal of DenseNet121 is to classify the input X-ray image into normal, pneumonia,
and COVID-19. The DenseNet121 acting as the attending physician extracts features
from the CXRs, and then this is used to fine-tune the DenseNet121 acting as the
Resident Doctors. The tuned DenseNet121 is used to train MobileNetV2 and
SqueezeNet using knowledge distillation. MobileNetV2 and SqueezeNet extract
features and predict whether the condition of the patient has improved, is stable, or
has deteriorated. Authors use probabilistic compact loss for training the
MobileNetV2 and SqueezeNet, and the following equation displays the same:
1X K X X K
Lpc ðqÞ ¼ max 0; fj ðxik Þ þ x fk ðxik Þ (10.3)
N k¼1 i 2S j¼1;j6¼k
k k
where x is the tuning parameter for the network, larger the value, more would be
the interclass compactness. MobileNetV2 and SqueezeNet were tested for their
resource consumption on six mobile phones. Overall, MobileNetV2 was better than
SqueezeNet for high performance phones, whereas SqueezeNet is more suitable for
low performance phones.
AlexNet
AlexNet model won ILSVRC10 and had a top-5 error of 17.0%. Maghdid et al. [32]
presented a comparison between modified AlexNet and a custom CNN by training
the model over CT scan and X-ray images. The custom model consists of an input
layer, and this was followed by a convolutional layer with 16 filters of size 5 5 and
ReLU activation function, following this was batch normalization, two fully con-
nected layers, and finally the output layer with two neurons and SoftMax as activa-
tion function. The weights were initialized using Glorot and cross-entropy as loss
function. The modified AlexNet network had the input size of 127 127 3, the
batch size was set to 10, trained for 20 epochs, the learning rate was initialized to
0.0003. The accuracy for X-ray was 94% and for CT scan was 94.1% on the custom
CNN, whereas an accuracy of CT scan was 82%, and X-ray was 98% for modified
AlexNet. A limitation to the work is that the proposed models were only trained for
20 epochs, and further increase in accuracy could be observed.
10.2.1.3 Ensemble
As we have seen, transfer learning models have performed individually well,
researchers have tried to combine these models together for even better results. One
such work was presented in [33], the authors’ ensemble MobileNetV2 and
SqueezeNet for feature extraction from the CXR images, and then classified into
normal, pneumonia, and COVID-19 using SVM classifier. They preprocessed the
image with the help of fuzzy colour technique. Next, these images were given as
input to MobileNetV2 and SqueezeNet and features are extracted keeping momen-
tum as 0.9, decay as 0.000001, batch size as 64, using stochastic gradient decent
activation function, and learning rate as 0.00001 for both the models. These features
were then clubbed using social mimic optimization and given as input to SVM for
classification. When MobileNetV2 and SqueezeNet were used individually using the
same method, the overall accuracies were 98.54% and 97.81%, respectively, whereas
the ensemble model gave an overall accuracy of 99.27% which is clearly better than
the individual models. The future work of this model, as stated by the authors, is to
use deep learning to find the impact of COVID-19 on other organs to.
Another work was presented in [34] by using DenseNet121 and VGG19 for
feature extraction from CXR images and then using SVM for classifying the fea-
tures into normal, COVID-19, and pneumonia and named this model as CoVNet-
19. In the DenseNet121 model the last fully connected layer was removed, the
model now gave an output of 32 features in the shape of 32 1, like DenseNet121,
VGG19 was also gave an output of 32 features in the shape of 32 1 after the
removal of the last fully connected layer. These features were then concatenated to
the shape 64 1 and used to train SVM. Before giving input to the model, the
images in the COVID-19 class are augmented. For VGG19 and DenseNet121, the
Adam optimizer was used to train the model with the learning rate of 0.001 and
decay rate of 10 after each epoch, and batch size of 32. Individually the training and
testing accuracy for three-class classification of VGG19 and DenseNet121 was
96.17% and 96.08%; and 97.79% and 96.30%, respectively, whereas the CoVNet-
19 gave testing and training accuracy as 98.28% and 99.02%. For two-class
classification, the training and testing accuracy of VGG19 and DenseNet121 was
98.47% and 99.61%; and 99.87% and 99.60%, respectively, whereas the CoVNet-
19 gave testing and training accuracy as 100% and 99.71%. Future work as stated
by the authors is to eliminate the limitations to their proposed model, creating a
lighter version of the model so that it can be deployed on mobile phones.
10.2.1.4 Generative networks

Generative networks have multiple application areas like generating cartoon,
recreating faces, text to image translation, and 3D object generation. Another appli-
cation area of generative networks is to augment datasets, Waheed et al. [36] pre-
sented such an application in the case of ACGAN (auxiliary classifier generative
adversarial networks) where it was used to produce synthetic X-ray images and
predicting COVID-19 and named the model as CovidGAN. Considering the limited
size of the dataset, authors realized the need of augmentation; therefore they used the
ACGAN approach. The generative network recreated images and the discriminative
network labels these images. The generative network took vector-containing noise
which was normally distributed and was random in nature, having a standard
deviation of 0.02 and the class label as inputs and obtained as single-output image of
size 112 112 3. The class labels were passed through the embedding layer, then
through a 7 7 node dense layer, have linear activation function giving an output of
7 7 1, and the noise tensor with initial shape of 1,024 7 7 gave an output of
7 7 1,024. The class label and the noise tensor are combined and then passed
through four convolutional layers, where the output shapes were 14 14 512 for
first layer, 28 28 256 for second layer, 56 56 128 for third layer, and finally
112 112 3 for the fourth and the final layer. The discriminative network took the
image of size 112 112 3 generated by the generative network and the original
image then classifies the image as real or face and assigns label of whether it is
COVID-19 or normal. Discriminative network had four convolutional layers, each
convolutional layer was followed by a batch normalization layer and a Leaky ReLU
activation layer; at the end we had a flatten layer and a dropout layer with the
threshold being 50%. The output size of the discriminative network is 56 56 64
for first layer, 28 28 128 for second layer, 14 14 256 for third layer, and
finally 7 7 512 for the last convolutional layer. This model was compared with
the performance of VGG16. VGG16 used an Adam optimizer for training with a
learning rate of 0.001, and the batch size was set to 16. The VGG16 model gave an
accuracy of 85%, sensitivity of 69%, and specificity of 95%, whereas the CovidGAN
model gave an accuracy of 95%, sensitivity of 90%, and specificity of 97%. As seen
from the results, CovidGAN outperformed the VGG16 model. The authors of the
paper tend to improve the GAN network of the model and work on a dataset that is
big and is from a single source, rather than using multiple sources.
10.2.1.5 Generalized regression neural network and

probabilistic neural network
GRNNs are improved neural networks based on nonparametric regression. It is an
advanced variant of radial basis neural network. PNN on the other hand is a
feed-forward neural network used for classification and is a variant of the Bayesian
Network. Ardhya et al. [37] proposed a model that combines the two approaches
using the AND operator. The proposed model works as a multi-class classifier. It
classifies CXR images into four classes – normal, pneumonia viral, pneumonia
bacterial, COVID-19. Three sets of experiments are covered in this study – binary
classification (COVID-19 vs normal), three-class classification (COVID-19 vs
normal vs pneumonia bacterial), and four-class classification (COVID-19 vs nor-
mal vs pneumonia bacterial vs pneumonia viral). The dataset used in this study is
created by Joseph Cohen. Since the dataset is very small, one-shot learning tech-
nique was used. To ensure maximum accuracy, the model is trained using all
images, one at a time. However, an optimum number of training iterations is fixed
empirically. It derives raw features from the image which are then fed to the GRNN
and PNN models. However, it is important to choose the best samples and form a
cluster using the enhanced features. Thus, there is a need to combine the results
obtained by these models using the ensemble AND operation. Once the clusters are
ready, the model can be tested using the testing samples. Raw features are extracted
from the testing sample and fed to the ensemble model which classifies it. For
experiment 1 (binary classification), 100% accuracy was achieved. In experiment 2
(three-class classification), the highest accuracy of 85.23% was achieved when
13 samples were taken into consideration. Finally, for the last experiment (four-
class classification), an accuracy of 74.05% was achieved when a number of
samples were 29. Although the model considerably fails for four-class classifica-
tion, it ensures a 100% accuracy in its binary classification.
10.2.1.6 YOLO
Bhuyan et al. [35] used segmentation techniques followed by applying deep
learning to detect COVID-19 in CXRs and CT scans. YOLO with full-resolution
CNN was used to segment the infected areas in the images. The authors used a
dataset consisting of 2,794 images for training and 1,061 images for testing. After
segmentation, a CNN is used to classify if it is COVID-19. Classification was
performed over 4-fold cross-validation of test dataset. In the fourth cross-fold
validation with mass segmentation, the sensitivity was 96.66%, specificity was
99.41%, accuracy was 99%, MCC was 96.07%, and F1 score of 96.66%.
10.2.1.7 Machine learning

We have seen how deep learning algorithms were used to detect COVID-19 from
CXRs, CT scans, and ultrasounds. Now let us have a look at the application of
machine learning algorithms for detecting COVID-19. Authors of [38] have pre-
sented us with the used of F-tree, iterative neighbourhood component feature selec-
tor, and machine learning algorithms like DT, LD, SVM, ensemble, and KNN for
detecting COVID-19. F-transform or fuzzy transform is used for image reduction.
Multi-kernel local binary pattern is used to extract features. Next iterative neigh-
bourhood component analysis (INCA) selects the best features. For applying INCA,
first min–max normalization is applied on the extracted features, then NCA weights
are generated from the normalised features, and finally with the help of KNN and
10-fold cross-validation, they extract the top 16 features. Using these features the
previously stated group of classifiers were trained. In the DT group, tree fine gave an
average accuracy of 83.57% and tree medium gave an average accuracy of 83.56%.
Linear discriminant gave an average accuracy of 79.11%, and in the ensemble
family, bagged tree gave an accuracy of 90.92%, subspace discriminant gave an
average accuracy of 89.87%, and subspace KNN gave an average accuracy of
95.58%. For SVM they used different kernels; for cubic kernel, the average accuracy
was 96.23%; for quadratic kernel, the average accuracy was 95.84%; for Gaussian
kernel, the average accuracy was 95.37%; and the average accuracy for linear kernel
was 94.75%. Similarly, different kernels were used for KNN also; for fine kernel, the
average accuracy was 95.19%; for medium kernel, the average accuracy was
92.25%; for coarse kernel, the average accuracy was 85.84%; for cosine kernel, the
average accuracy was 91.70%; for cubic kernel, the average accuracy was 91.79%;
and the average accuracy for weighted kernel was 94.78%. As seen from the results,
SVM with cubic kernel gave maximum average accuracy which was 96.23%. The
limitation to this study is the small size of the dataset. Since they have not used many
images, so their model would not be generalized.
Another study was performed in [39] to use local binary patterns and machine
learning algorithms to detect COVID-19. Local binary pattern is a method of tex-
ture feature extraction of images. The pixel values are compared with that of its
neighbours and a binary string is generated using the difference. This binary string
is converted to decimal, and this becomes the value of that pixel. The hyperpara-
meter for this is the radius (R) and the number of neighbours (P). They then applied
the extracted features to train naı̈ve GNB, KNN, SVM, random tree, and RF; they
also used ensemble of RF-RT-SVM, RF-RT-KNN, and RF-SVM-KNN. They
applied different combinations of R and P on the model are (1,8), (2,16), (3,24),
(4,32), (5,40), (6,48), (7,56), (8,64), (9,72), and (10,80); the average accuracies for
the models on these combinations are 84.467%, 83.525%, 84.344%, 83.525%,
83.361%, 83.852%, 83.648%, 81.926%, 80.328%, and 80.328%, respectively. It is
evidently visible that R1 (1,8) gave the best average accuracy, and overall, the
ensemble of RT-RF-KNN for (6,48) and (7,56) gave the best accuracy as 89.180%.
The limitation of this study is the small sample size used for the study.
Thepade et al. [40] proposed a way of detecting COVID-19 with the help of
global texture feature extraction using colour space. The images are converted from
RGB (red green blue) to YCrCB, Kekre-LUV, and CIE-LUV colour spaces, a feature
vector is generated considering different combinations of colour space. The first
feature vector is the concatenation of CIE-LUV and YCrCb, the second vector is the
concatenation of CIE-L, Y, and Kekre-LUV, and the third vector is the concatenation
of Kekre-LUV and YCrCb. These feature vectors are given as input to machine
learning model for classification; the models used were RF, ET, GNB, and ensemble
of ETþRFþSimpleLogistic (SL), ETþRFþGNB, and ETþRFþRT. These models
were trained using 10-fold cross-validation. For the first feature vector (YþCIE-
LþKekre-LUV), GNB gave an accuracy of 65%, ET gave an accuracy of 71.67%,
RF gave an accuracy of 82.92%, whereas the ensemble models ETþRFþRT,
ETþRFþNB, and ETþRFþSL gave accuracies of 80.83%, 80.83%, and 84.167%,
respectively. Clearly, the ensemble of ETþRFþSL performed best for this feature.
When the second feature vector (YCrCbþKekre-LUV) was used, GNB gave an
accuracy of 66.67%, ET gave an accuracy of 78.33%, RF gave an accuracy of
82.917%, whereas the accuracies of the ensemble models ETþRFþRT,
ETþRFþGNB, and ETþRFþSL were 80.83%, 85.83%, and 87.083%. Again, the
ensemble of ETþRFþSL gave the best accuracy for the second feature vector too.
For the third feature vector (YCrCbþCIE-LUV), the accuracy for GNB, ET, and RF
were 67.5%, 80.83%, and 84.583%, respectively, whereas for the ensemble models
of ETþRFþRT, ETþRFþGNB, and ETþRFþSL the accuracies were 83.33%,
85%, and 84.167%, respectively. For the third feature the ensemble of ETþRFþNB
gave the best results. Overall, the combination of the feature vector YCrCbþKekre-
LUV and ETþRFþSL gave the best results of 87.083% accuracy, 87% precision,
87.1% recall, 86.9% F1 score, and 79.5% Matthews correlation coefficient. Though
the proposed model has low computational power, but the data used to train the
model is less, due to which the results are not up to the mark.
10.3 Comparative analysis of existing work
In the previous section, we have reviewed major research works in COVID-19

detection and classification using machine learning and deep learning approaches;
however, a comparative analysis of these works in the perspective of dataset used,
augmentation techniques, major contributions, and challenges will throw light for
future research. In Figure 10.2, we have presented the accuracy comparison
between the models used in the reviewed works.
In Figure 10.2 we can see the comparison on accuracy between multi-class
classification (two class, three class, and four class). For two class, Generative
networks [24] give us the best accuracy of 100% and an ensemble of DenseNet and
VGG gives an accuracy of 99%. Similarly for three-class classification, the best
results are shown by VGG and SqueezeNet, both giving an accuracy of 96%. In the
case of four-class classification, GoogLeNet gave an accuracy of 95%, and CNN
gave an accuracy of 91%.
In Table 10.1, the major limitations of the research works identified were
unavailability of large datasets, imbalanced datasets, and the use of multiple image
Accuracy comparison
100 99 98
96 989898
99 989999
95
100
95 94
98 98
9594 95
CNN(7)
91 StackNet(10)
89 87
85 GoogleNet(11)
80 VGG(15)
80 ResNet(17)
DenseNet(21)
Accuracy
SqueezeNet(24
60 Xception(25)
MobileNetV2(28)
AlexNet(30)
40 MobileNetV2 and SqueezeNet(31)
DenseNet and VGG(32)
Generative Networks(33)
GRNN and PNN(34)
20 SVM(35)
KNN(36)
LR(37)
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0
Two class Three class Four class
Models (research work)
Figure 10.2 Accuracy comparison of state-of-the-art machine learning and deep

learning algorithms
Table 10.1 Comparative study of papers reviewed
Research Feature ex- Dataset Sample size and classes Data augmenta- Class imbal- Major contri- Limitations
work traction tion ance butions
[12] Yes, using COVID-19 chest COVID-19: 634; pneu- Yes, Synthetic Yes, solved Proposed
VGG16 X-ray dataset monia: 4,474; normal: minority using aug- CovStacknet
(GitHub), 1,583; 2 combination of oversampling mentation
COVID-19 chest classes technique
X-ray (Kaggle), Normal and pneumo-
and labelled OCT nia; COVID-19 and
and chest X-ray non-COVID-19
images for clas-
sification (Kag-
gle)
[36] No IEEE COVID- COVID-19: 403; normal: Yes, using Yes Proposed a state GAN architecture
19 chest X-ray 721 ACGAN of art Covid- can be improved
dataset, COVID- GAN further; small da-
19 radiography taset
database,
COVID-19 chest
X-ray dataset
initiative
[37] Yes, using one Joseph Cohen 306 (69 COVID-19, 79 in No No 100% accuracy Accuracy of four-
shot learn- (COVID-19 chest normal, pneumonia in binary class classifica-
ing X-ray dataset) viral, pneumonia classification, tion is very low
bacterial); 3 combina- one-shot
tion of classes COVID- learning
19 and normal; normal, approach on
bacterial pneumonia, COVID-19
and COVID-19; dataset
normal, bacterial pneu-
monia, viral pneumo-
nia, and COVID-19
[9] No COVID-R 2,843 COVID-19 images, No Yes Proposed a cus- Model performed
3,108 normal images tom made 22- poorly in low
and 1,439 pneumonia layer CNN quality X-ray
images: a total of 7,390; images
COVID-19 and normal;
monia, and COVID-19;
monia, viral pneumo-
nia, and COVID-19
[27] No SIRM, Stanford 288 samples; normal and Yes, Python No Proposed a cus- Too many false
CheXpert COVID-19; normal, OpenCV tom CNN predictions
COVID-19, and pneu- model
monia
[11] No Paediatric CXR da- Normal: 7,595; unknown No Yes 22 High class imbal-
taset, RSNA pneumonia: 6,012; bac- ance
CXR dataset, terial pneumonia:
twitter COVID- 2,780; COVID-19: 313;
19 CXR dataset, normal and COVID-19;
Montreal normal, bacterial pneu-
COVID-19 CXR monia, and COVID-19
dataset
[13] No Italian society of 240 images out of which No Yes Proposed a fu- There are 120
medical and in- 120 were for COVID- sion model of images for
ternational radi- 19; COVID-19 and GoogLeNet COVID-19 due
ology and other non-COVID-19; and random to which it leads
medical re- COVID-19, normal, forest for im- to a bias dataset
sources pneumonia, and H1N1 age and text
classification
(Continues)
[14] No COVID-19 image 142 PA images from 192 Yes, rotate No Proposed a deep The dataset is very
data collection images COVID-19 and images, hori- learning based small in size
and Kaggle’s 142 from the normal zontal and ver- neural net-
chest X-ray class of the Kaggle’s tical flip work called
images (pneumo- chest X-ray images da- nCoVNet
nia) taset; 2 classes COVID-
19 and non-COVID-19
[15] No CT-scan: COVID- X-ray: 729; CT: 746; ul- Yes No Made a compar- –
19-CT-dataset; trasound: 911; X-ray ison between
ultrasound: PO- classes-normal, all popular
COVID-Net: COVID-19, and pneu- imaging tech-
(POCUS); X- monia; CT classes- niques as well
rays: COVID-19 normal and COVID-19; as popular
image data col- ultrasound classes- CNN models
lection: prospec- normal, COVID-19,
tive predictions and pneumonia
are the future
[16] No COVID images: 181 COVID-19, 364 nor- No Yes High accuracy Research based on
publicly avail- mal with just 1 literature review
able papers/arti- false negative that may not be
cle, normal: NIH peer-reviewed
clinical centre
[17] No Dataset collected 6,523 images out of which Yes (rotation by Yes Highlighting the The dataset is
from various 250 were for COVID- 15 degrees) areas of in- highly imbal-
medical facilities 19 and 3,520 were nor- fection in the anced, study was
mal chest X-rays restricted to only
one model
[18] No COVID chest X-ray 406 images (203 in Rotated by an an- No Compared most Models are not
dataset for COV- COVID-19 and normal gle of 5 degrees popular mod- tested for multi-
ID images, classes each) (clockwise), els with var- class classifica-
GitHub repo horizontally ious tuneable tion
chest X-ray 8 da- flipped, scaled hyperparame-
taset for normal by 15% and ters
Gaussian noise
(mean ¼ 0 and
variance¼0.25)
[19] No Dataset collected COVID-19: 1,296; pneu- No No Highlighting the The dataset is small,
from 6 medical monia: 1,735; normal: infected area the model was
facilities 1,325 using grad- not analysed over
cam accuracy, preci-
sion, and
F1 score
[20] Yes, Re- COVID-19 X-ray 180 COVID-19, 200 nor- randomly rotat- No Proposed and –
sNet18, Re- chest and CT – mal ing, shifting, compared
sNet50, Re- Kaggle; and flipping the three different
sNet101, X-ray chest – training images ways of de-
VGG16, normal images tecting
and VGG19 COVID-19
[21] Yes, Re- COVIDx, 5 hospi- 4,119 COVID-19, 8,173 No Yes Used a combi- Achieved accuracy
sNet50 tals from partners control, 5,763 non- nation of deep not satisfactory
healthcare system COVID-19 pneumonia learning and for medical based
in the United machine research
States and 4 hos- learning
pitals in South
Korea
(Continues)
[22] No Custom dataset, 5,184 (5,000 normal, 184 Yes; Flipping, Yes Achieved sensi- Data was imbal-
COVID-19-X- COVID-19) small rotation, tivity of ap- anced even after
ray-5k, using adding small prox. 98% for augmentation
COVID-19 chest amount of dis- all models
X-ray dataset and tortions
CheXpert
[23] Yes, using 12 COVID-19 images Both datasets: 388 images Yes; Affine trans- No Performed a to- Limited dataset for
CNN mod- – JP Cohen, (194 in COVID-19 and formations tal of 144 ex- training DL
els COVID-19-X- non-COVID-19 classes periments models
rays (Kaggle) each) using different
Normal (Dataset combinations
A) – chestX-ray of CNN mod-
images (pneumo- els and classi-
nia)Normal (Da- fiers
taset B) – NIH
chest X-rays
[24] Yes; Dense- COVID-19-CT- 746 images (397 non- Yes; reflection No Used extreme Data augmentation
Net201 ar- dataset COVID-19, 349 COV- (relative to the learning ma- was overused
chitecture ID-19) x and y axes) chine (EML) (dataset increased
rotation (angle from 746 to
values of 90 3,730)
and 270 de-
grees)
[25] Yes; Re- Dataset 1: chest X- Dataset 1: 5,856 (bacteria, Yes; Affine trans- Yes Model devel- Difference in image
sNet50, Ray images virus, normal) formation, his- oped gives sizes led to addi-
Dense- (pneumonia) Dataset 2: 441 (COV- togram en- higher accu- tional steps
Net169 Dataset 2: COV- ID-19, other viruses) hancement racy as com-
ID-19 chest X- Dataset 3: 2,000 (1,000 pared to base
ray dataset images, 1,000 masks) models
(GitHub)
Dataset 3: U-Net
[26] No Italian society of COVID-19: 190; normal: Yes, rotation, Yes, solved Proposed the ef- Lower number of
medical and in- 1,341; viral pneumonia: scaling, image by image ficiency of images in the da-
terventional radi- 1,345 translation augmenta- SqueezeNet taset
ology COVID-19 tion over Dense-
database, novel Net201, Re-
corona virus sNet18, and
2019 dataset, AlexNet
COVID-19 posi-
tive chest X-ray
images from dif-
ferent articles,
chest X-ray
images
[27] Na COVID-19 images 1,300 (290 COVID-19, No No Proposed model Proposed model
– GitHub repo by 310 normal, 330 bac- is computa- needs further
Joseph; normal, terial pneumonia, 327 tionally less clinical testing
pneumonia bac- viral pneumonia) expensive and
terial, pneumonia gives 99% ac-
viral – Kaggle’s curacy for
chest X-ray two-class
images (pneumo- classification
nia)
[28] No Chest X-ray (COV- Total ¼ 6,432; Yes, random rota- Yes Xception gave More models could
ID-19 and pneu- train ¼ 5,467 (COVID- tion 100% training have been com-
monia) 19 ¼ 490; (range ¼ 10 accuracy and pared
normal ¼ 1,345; degrees), 97% testing
pneumonia ¼ 3,632) flipped hori- accuracy
zontally,
zoomed in or
out
(range ¼ 0.4)
(Continues)
[29] No COVID-19 radio- 2,905 images (219 No Yes 100% sensitivity More popular CNN
graphy COVID-19, 1,341 nor- for COVID- models could
mal, 1,345 pneumonia) 19 class using have been com-
Xception pared
[30] No (1) GitHub repo by 1,428 (224 COVID-19, No Yes Comparative Study does not
Cohen; (2) Kag- 700 pneumonia, 504 study of all consider other si-
gle – RSNA, normal) recent and milar virus that
radiopaedia, trending mod- may result in si-
SIRM; (3) (Ker- els – best per- milar X-rays
many et al.) formance by
MobileNetV2
[31] No RSNA pneumonia DS1: 179 images for nor- No No Develop a mo- Authors do not
detection chal- mal, pneumonia, and bile app consider the tra-
lenge (DS1) and COVID-19 classes COVID- ditional perfor-
COVID-19 im- each; DS2: 76 worse, MobileXpert mance evaluators
age data collec- 38 stable, and 45 im- like accuracy and
tion (DS2) proves loss
[32] No Chest X-ray images X-ray: COVID-19: 85 and No Yes Comparison of The dataset size is
(pneumonia) normal: 85; CT scan: transfer learn- small, authors
(Kaggle), radio COVID-19: 203 and ing and cus- have not ex-
helia, COVID- normal: 153 tom CNN plored other
19 chest X-ray CNNs
dataset (GitHub),
BSTI
[33] Yes, Mobile- Joseph Paul Cohen 458 (295 COVID-19, 65 No Yes Proposed an en- –
NetV2 and dataset, Qatar normal X-rays, 98 semble of
SqueezeNet University (Kag- pneumonia) MobileNetV2
gle – X-ray) and Squeeze-
Net for fea-
ture extraction
[34] Yes, VGG19 COVID-19 radio- COVID-19: 798; normal: Yes, width shift, Yes, tried to Proposed a state The dataset still re-
and Dense- graphy database 2,341; pneumonia: height shift, solve it of art model mains imbal-
Net201 (Kaggle), chest 2,345 zooming, using aug- called anced after
X-ray images shearing, and mentation CoVNet-19 augmentation.
(pneumonia) rotation by a Authors should
(Kaggle), COV- small angle have taken the
ID-19 chest X- same number of
ray dataset neutral and
(GitHub), COV- pneumonia
ID-19 chest X- images and then
ray dataset initia- applied augmen-
tive (GitHub), tation to the en-
and COVID-19 tire dataset
X-ray dataset
(Kaggle)
[38] Yes, using GitHub, Kaggle COVID-19: 135; pneu- No No Proposed the de- The dataset used is
multi- monia: 150; normal: tection of merely 435
kernel local 150 COVID-19 images, this is
binary pat- using machine very less to cre-
tern learning and ate a generalised
F-transform model for practi-
cal use
[39] Local binary COVID-19 image 305 (68 COVID-19, 79 No No Complete ma- Overall accuracy is
patterns data collection normal, 158 pneumo- chine learning higher for DL
(LBP) nia) based model models
(Continues)
[40] Yes, YCbCr, COVID-19 image 60-COVID-19; 60-nor- No No Use of lumi- Small dataset; max-
Kekre- data collection mal; 60-viral pneumo- nance Chroma imum accuracy
LUV, and nia; 60-bacterial features for produced is of
CIE-LUV pneumonia COVID- 84.583% which
19 classifica- is low for medi-
tion cal applications,
hence model
cannot be de-
ployed in real
time
[41] Yes, non- – COVID-19: 291; influen- No No Random forest Small dataset, ac-
subsampled za A/B: 160; other dis- based model curacy is not
dual-tree eases: 119 for detection good for medical
complex COVID-19 application
contour let from other
transform pneumonia
and grey images
level co-
occurrence
sources. We have tried to address these limitations in the next section and provided
a way for working ahead on these challenges.
10.4 Research gaps

Although most papers were successful in designing models that gave a great
accuracy score, there were certain common limitations and research gaps. The
following are the most important gaps.
10.4.1 Unavailability of large datasets

The sample size of the datasets for most of the research works was very small. The
size for the COVID-19 class was even smaller. Since COVID-19 has spread to most
of the countries, there have been multiple variants in the COVID-19 virus. There is
a need of a bigger dataset, to be able to generalize the model for detecting of
COVID-19. A dataset which has samples from different age groups and different
locations could help us in developing and training a model that can be generalized
and be capable of deployment. For creating such a large dataset, we need the col-
laboration of local hospitals and the respective state and centre governments,
without which collecting data at such a large scale is difficult. In [36] the authors
have mentioned about the challenges involved in having a small dataset. They used
Generative Adversarial Networks (GAN) to generate synthetic images to increase
the sample size. In future, researchers can enhance existing approaches for devel-
oping synthetic dataset and can also come up with unique and efficient ways.
10.4.2 Imbalanced datasets

Imbalanced dataset is one of the most prevalent problems that affect the efficient
building of machine learning models. In most of research work for COVID-19
detection, authors have addressed the problem of imbalanced datasets. The COVID-
R dataset used by researchers in [9] have 3,108 normal images and 4,248 anomalous
images out of which 2,843 are COVID-19 images and 1,439 are pneumonia. A
similar sample ratio of 1:2 (overall healthy to COVID-19) is maintained in [16] as
well. The COVID-CXR dataset that is used in [36] has a total of 1,124 CXR images
(403 in COVID class and 721 in normal class). However, the authors have generated
synthetic datasets to compensate for the small and imbalanced dataset. In [17] the
authors have used the 6,523 CXRs where 250 of them are of COVID-19, 2,753 of
patients with other pulmonary diseases and 3,520 healthy patients. In this case the
COVID-19 samples are very less compared to other class samples. This case shows a
highly imbalanced dataset where the COVID-19 samples are very less in proportion
to healthy and other pulmonary diseases. In Table 10.1 the research works that have
used imbalanced datasets are mentioned and most of them have used synthetic
samples or weighted learning approaches to address this issue. However, the
approaches should be verified against all performance measures and real-time diag-
nosis with radiologists. Furthermore, the computational complexity of these
approaches should be tested against the modified samples.
10.4.3 Multiple image sources

In the research works we have analysed; the authors have used datasets from
multiple sources. In research work [12], the dataset is developed by combining data
from GitHub and Kaggle. The COVID-CXR images are combined from GitHub
and labelled OCT and CXR images for classification from Kaggle. In [22] the
authors have used COVIDx and data collected from few hospitals in the United
States and South Korea. Similarly, from Table 10.1, it is evident that most of the
research works have used datasets from varied sources. It is imperative to analyse
the extensibility and generalizability of the developed works. The challenges
involved in datasets are due to the varying characteristics of demographics, dif-
ferent age-groups, history of medical conditions, lifestyle disorders, and various
physiological properties across sex, race, and ethnic groups. Decision-making in
medical diagnosis should accommodate all the complexities involved. Hence there
is a need for validation of models developed with one dataset to be tested with other
datasets.
10.5 Conclusion
With COVID-19 spreading all over the world, it has become vital that we try to
detect the virus at the earliest so that the infected person can be isolated and treated.
However, the most popular approach to detect the virus is the RT-PCR test that
takes hours to give results. The next best shot at detecting the virus is using
radiography techniques like CXRs, CT-scans, and ultrasounds. Although the ima-
ging is done within a few minutes, it requires experts and professionals to analyse
the obtained images and to detect the virus. Possibility of human errors increases
the misinterpretation rates. Thus, it is vital to develop a system that not only makes
such predictions in seconds but also yields satisfactory results. In this chapter, we
have summarized various research studies based on machine learning and deep
learning models. Majority of the models are trained using real-time data and yield
good results. Though training these models may be time-consuming, once trained,
they take a few milliseconds to test each sample. However, as discussed in many
papers, the biggest disadvantage to any COVID-19 related study is that there is very
limited amount of reliable data which is not sufficient to train deep learning mod-
els. Although the chapter features some papers that give high accuracies, the idea of
the model being biased due to imbalanced datasets cannot be completely elimi-
nated. Thus, future research should be done on datasets that have a sufficient
number of samples in each class. The best way to tackle the current problem is to
merge different models for detecting COVID-19. Merging the X-ray or CT scan
method with supervised machine learning models that detect COVID-19 using
various parameters like age, sex, diabetic, pneumonia, as discussed in [42], can
result in error-free and more accurate results. Apart from clinical methods to limit
the spread of the virus, various non-pharmaceutical techniques can be used as well.
One such technique as mentioned in [43] is to track the people who were in contact
with a positive patient. Thus, the most efficient way to fight the disease is to make a
hybrid system that combines various models to detect COVID-19 using radio
imaging techniques.
References
[1] Apollo Pulmonologist. New COVID-19 symptoms to look out for [Internet].
Ask Apollo; 2021. [cited 2021 May 5]. Available from: https://healthlibrary.
askapollo.com/new-covid-19-symptoms-to-look-out-for/.
[2] World Health Organization. Coronavirus disease (COVID-19): How is it
transmitted? [Internet]. World Health Organization. 2020 [cited2021 May 6].
Available from: https://www.who.int/news-room/q-a-detail/coronavirus-dis-
ease-covid-19-how-is-it-transmitted.
[3] Greenhalgh T, Jimenez JL, Prather KA, Tufekci Z, Fisman D and Schooley
R. Ten scientific reasons in support of airborne transmission of SARS-CoV-
2. Lancet. 2021; 397(10285): 1603–5.
[4] Rahman A, Chakraborty C, Anwar A, et al. SDN–IoT empowered intelligent
framework for industry 4.0 applications during COVID-19 pandemic.
Cluster Comput. 2021; 4. Available from: https://doi.org/10.1007/s10586-
021-03367-4.
[5] Dash S, Chakraborty C, Giri SK and Pani SK. Intelligent computing on time-
series data analysis and prediction of COVID-19 pandemics. Pattern
Recognit Lett [Internet]. 2021; 151: 69–75. Available from: https://doi.org/
10.1016/j.patrec.2021.07.027.
[6] Scohy A, Anantharajah A, Bodéus M, Kabamba-Mukadi B, Verroken A and
Rodriguez-Villalobos H. Low performance of rapid antigen detection test as
frontline testing for COVID-19 diagnosis. J Clin Virol. 2020; 129.
[7] Chendrasekhar A. Chest CT versus RT-PCR for diagnostic accuracy of
COVID-19 detection: a meta-analysis. J Vasc Med Surg. 2020; 8(4): 1–4.
[8] Yasin R and Gouda W. Chest X-ray findings monitoring COVID-19 disease
course and severity. Egypt J Radiol Nucl Med. 2020; 51(1).
[9] Hussain E, Hasan M, Rahman MA, Lee I, Tamanna T and Parvez MZ.
CoroDet: A deep learning based classification for COVID-19 detection
using chest X-ray images. Chaos, Solitons Fractals [Internet]. 2021; 142:
110495. Available from: https://doi.org/10.1016/j.chaos.2020.110495.
[10] Hernandez D, Pereira R and Georgevia P. COVID-19 detection through X-
ray chest images. 2020 Int Conf Autom Informatics, ICAI 2020 – Proc. 2020.
[11] Rajaraman S, Siegelman J, Alderson PO, Folio LS, Folio LR and Antani SK.
Iteratively pruned deep learning ensembles for COVID-19 detection in chest
X-rays. IEEE Access. 2020; 8: 115041–50.
[12] Rabbah J, Ridouani M and Hassouni L. A new classification model based on
StackNet and deep learning for fast detection of COVID 19 through X rays
images. 4th Int Conf Intell Comput Data Sci ICDS 2020. 2020.
[13] Mayya A and Khozama S. A novel medical support deep learning fusion
model for the diagnosis of COVID-19. Proc IEEE Int Conf Advent Trends
Multidiscip Res Innov ICATMRI 2020. 2020; 2–7.
[14] Panwar H, Gupta PK, Siddiqui MK, Morales-Menendez R and Singh V.

Application of deep learning for fast detection of COVID-19 in X-Rays
using nCOVnet. Chaos, Solitons Fractals [Internet]. 2020; 138: 109944.
Available from: https://doi.org/10.1016/j.chaos.2020.109944.
[15] Horry MJ, Chakraborty S, Paul M, et al. COVID-19 detection through
transfer learning using multimodal imaging data. IEEE Access. 2020; 8:
149808–24.
[16] Vaid S, Kalantar R and Bhandari M. Deep learning COVID-19 detection bias:
accuracy through artificial intelligence. Int Orthop. 2020; 44(8): 1539–42.
[17] Brunese L, Mercaldo F, Reginelli A and Santone A. Explainable deep
learning for pulmonary disease and coronavirus COVID-19 detection from
X-rays. Comput Methods Programs Biomed [Internet]. 2020; 196: 105608.
Available from: https://doi.org/10.1016/j.cmpb.2020.105608.
[18] Minaee S, Kafieh R, Sonka M, Yazdani S and Jamalipour Soufi G. Deep-
COVID: Predicting COVID-19 from chest X-ray images using deep transfer
learning. Med Image Anal. 2020; 65.
[19] Nayak SR, Nayak DR, Sinha U, Arora V and Pachori RB. Application of deep
learning techniques for detection of COVID-19 cases using chest X-ray ima-
ges: A comprehensive study. Biomed Signal Process Control [Internet]. 2021;
64: 102365. Available from: https://doi.org/10.1016/j.bspc.2020.102365.
[20] Li L, Qin L, Xu Z, et al. Artificial intelligence distinguishes COVID-19 from
community acquired pneumonia on chest CT. Radiology [Internet]. 2020;
Available from: http://arxiv.org/abs/2003.13865.
[21] Ismael AM and Şengür A. Deep learning approaches for COVID-19 detec-
tion based on chest X-ray images. Expert Syst Appl. 2021; 164.
[22] Zhong A, Li X, Wu D, et al. Deep metric learning-based image retrieval
system for chest radiograph and its clinical applications in COVID-19. Med
Image Anal [Internet]. 2021; 70: 101993. Available from: https://doi.org/10.
1016/j.media.2021.101993.
[23] Ohata EF, Bezerra GM, das Chagas JVS, et al. Automatic detection of
COVID-19 infection using chest X-ray images through transfer learning.
IEEE/CAA J Autom Sin. 2021; 8(1): 239–48.
[24] Turkoglu M. COVID-19 detection system using chest CT images and mul-
tiple kernels-extreme learning machine based on deep neural network. IRBM
[Internet]. 2021; 1: 1–8. Available from: https://doi.org/10.1016/j.irbm.2021.
01.004.
[25] Lv D, Qi W, Li Y, Sun L and Wang Y. A cascade network for Detecting
COVID-19 using chest x-rays. 2020; 1–22. Available from: http://arxiv.org/
abs/2005.01468.
[26] Chowdhury MEH, Rahman T, Khandakar A, et al. Can AI help in screening
viral and COVID-19 pneumonia? IEEE Access, 2020; 8: 132665–132676.
doi: 10.1109/ACCESS.2020.3010287
[27] Khan AI, Shah JL and Bhat MM. CoroNet: A deep neural network for
detection and diagnosis of COVID-19 from chest x-ray images. Comput
Methods Programs Biomed [Internet]. 2020; 196: 105581. Available from:

https://doi.org/10.1016/j.cmpb.2020.105581.
[28] Jain R, Gupta M, Taneja S and Hemanth DJ. Deep learning based detection
and analysis of COVID-19 on chest X-ray images. Appl Intell [Internet].
2021; 51(3): 1690–700. Available from: http://link.springer.com/10.1007/
s10489-020-01902-1.
[29] El Gannour O, Hamida S, Cherradi B, Raihani A and Moujahid H.
Performance evaluation of transfer learning technique for automatic detec-
tion of patients with COVID-19 on X-ray images. 2020 IEEE 2nd Int Conf
Electron Control Optim Comput Sci ICECOCS 2020. 2020; (1): 1–6.
[30] Apostolopoulos ID and Mpesiana TA. Covid-19: automatic detection from
X-ray images utilizing transfer learning with convolutional neural networks.
Phys Eng Sci Med [Internet]. 2020; 43(2): 635–40. Available from: https://
doi.org/10.1007/s13246-020-00865-4.
[31] Li X, Li C and Zhu D. COVID-MobileXpert: On-device COVID-19 patient
triage and follow-up using chest X-rays. Proc – 2020 IEEE Int Conf
Bioinforma Biomed BIBM 2020. 2020; 1063–7.
[32] Maghdid HS, Asaad AT, Ghafoor KZ, Sadiq AS and Khan MK. Diagnosing
COVID-19 pneumonia from x-ray and CT images using deep learning and
transfer learning algorithms. arXiv. 2020; 1–8.
[33] Toğaçar M, Ergen B and Cömert Z. COVID-19 detection using deep learning
models to exploit Social Mimic Optimization and structured chest X-ray images
using fuzzy color and stacking approaches. Comput Biol Med. 2020; 121.
[34] Kedia P, Anjum and Katarya R. CoVNet-19: A deep learning model for the
detection and analysis of COVID-19 patients. Appl Soft Comput [Internet].
2021; 104: 107184. Available from: https://doi.org/10.1016/j.asoc.2021.
107184.
[35] Bhuyan HK, Chakraborty C, Shelke Y and Pani SK. COVID-19 diagnosis
system by deep learning approaches. Expert Syst. 2021: 1–18. https://doi.
org/10.1111/exsy.12776.
[36] Waheed A, Goyal M, Gupta D, Khanna A, Al-Turjman F and Pinheiro PR.
CovidGAN: Data augmentation using auxiliary classifier GAN for improved
Covid-19 detection. IEEE Access. 2020; 8: 91916–23.
[37] Aradhya VNM, Mahmud M, Guru DS, Agarwal B and Kaiser MS. One-shot
cluster-based approach for the detection of COVID-19 from chest X-ray
images. Cognit Comput. 2021; 1–9.
[38] Tuncer T, Ozyurt F, Dogan S and Subasi A. A novel Covid-19 and pneu-
monia classification method based on F-transform. Chemom Intell Lab Syst
[Internet]. 2021; 210: 104256. Available from: https://doi.org/10.1016/j.
chemolab.2021.104256.
[39] Thepade SD and Jadhav K. Covid19 identification from chest x-ray images
using local binary patterns with assorted machine learning classifiers. 2020
IEEE Bombay Sect Signat Conf IBSSC 2020. 2020; 46–51.
[40] Thepade SD, Chaudhari PR, Dindorkar MR and Bang SV. Covid-19 iden-
tification using machine learning classifiers with histogram of luminance
Chroma features of chest x-ray images. 2020 IEEE Bombay Sect Signat Conf
IBSSC 2020. 2020; 36–41.
[41] Wu Z, Li L, Jin R, et al. Texture feature-based machine learning classifier
could assist in the diagnosis of COVID-19. Eur J Radiol. 2021; 137.
Mohammed IA. Supervised machine learning models for prediction of
COVID-19 infection using epidemiology dataset. SN Comput Sci [Internet].
2021; 2(1): 1–13. Available from: https://doi.org/10.1007/s42979-020-
00394-7.
[43] Garg L, Chukwu E, Nasser N, Chakraborty C and Garg G. Anonymity pre-
serving IoT-based COVID-19 and other infectious disease contact tracing
model. IEEE Access. 2020; 8: 159402–14.
Chapter 11
Machine-based drug design to inhibit
SARS-CoV-2 virus
T. Lurthu Pushparaj1, E. Francy Irudaya Rani2,
E. Fantin Irudaya Raj3 and M. Appadurai4
Abstract
Researchers worldwide are striving hard to design the best anti-coronavirus drug to
overcome the current pandemic situation. Due to the lack of animal and human
trials data, the process is still complicating and causing more death worldwide.
Since biological testing costs more money and time-consuming, the combination of
computerized programs-based evaluation like molecular docking, virtual screening
(AutoDock, HEX, Schrodinger, Gaussian, and Glide), and molecular dynamics
study serves as a hopeful way in drug developing and studying their effect over
severe acute respiratory syndrome (SARS)-coronavirus-causing components like
spike (S) protein present around the SARS-coronavirus-2 (SARS-CoV-2), main
protease (M-pro), and ribonucleic acid cooperate. The keen analysis of the amino
acid sequences in the coronavirus-19 infection-causing proteins will give very
important information about the virus transformation and replication cycle. The
amino acid sequences and their active sites provided in the 3D crystalline structure
of M-pro (PDB ID; 6LU7) afford valuable data to the researchers about the type of
inhibitors that corresponds to the SARS-CoV-2 inhibition. Compared to all other
deadly viruses like flu, human immunodeficiency virus, and SARS, novel cor-
onavirus SARS-CoV-2 shows superior binding affinity over a human transmem-
brane protein christened angiotensin-converting enzyme 2, found in human lungs.
Since the target is very clear, every scientist aims to design a new drug or check the
available prodrug activity through a computer program to defeat the COVID-19
disease. Nowadays, many synthetic and natural drugs have been tested for their
suitability against M-pro. Since the coronavirus-19 infection spreads more
1
PG Chemistry & MRI Research Lab, TDMNS College, Affiliated to MS University, Tirunelveli, India
2
Department of Electronics and Communication Engineering, Francis Xavier Engineering College,
Tirunelveli, Affiliated to Anna University, India
3
Department of Electrical and Electronics Engineering, Dr. Sivanthi Aditanar College of Engineering,
Tiruchendur, Affiliated to Anna University, India
4
Department of Mechanical Engineering, Dr. Sivanthi Aditanar College of Engineering, Tiruchendur,
Affiliated to Anna University, India
vigorously, the traditional real-time PCR test will need more time for infection
confirmation, so machine-based imaging studies like MRI, computerized tomo-
graphy (CT), and X-ray are needed. The MRI and CT need contrast agents (CAs) to
give more precise images. The development of image contrast-enhancing agents
will give more appropriate outcome image in detecting COVID-19 infection in the
early stages. The designing of perfect multimodal CAs is the current research
among MRI researchers. It will behave as both targeting and coronavirus-19 killing
drug for the current pandemic situation. All the above-said applications can be
accomplished only by designing the drug and then studying their binding studies
through a computer application.
Keywords: SARS-coronavirus-19; COVID-19 drug; ACE2 enzyme; Main

protease; Antiviral docking; Dual-MRI agent
11.1 Introduction
In the fast genetic research world, silicon chip-engineered devices have become a
right hand for researchers. Since the application of this is unlimited, a good science
researcher can develop a new genetically modified microbe, vaccine, or drug for
the deadliest disease. This becomes true on seeing the current pandemic situation
rooted by the deadliest beta microbes called severe acute respiratory syndrome
coronavirus-2 (SARS-CoV-2), a perfectly laboratory-made virus-containing BAT
and PIG genes [1]. Typically, viruses that reason illnesses in people are known to
program at least one protease that assumes significant parts in the viral life cycle.
Reacting rapidly to obscure microbes is essential to stop the uncontrolled spread of
sicknesses that lead to an outbreak, for example, the novel coronavirus. In the past
two decades, six coronaviruses, like SARS-coronavirus (fatality rates of 9.6%),
MERS-coronavirus (fatality 34%), HCoronavirus-HKU1, HCoronavirus-OC43,
HCoronavirus-NL63, and HCoronavirus-229E, have infected the human population
in the years 2002 and 2012.
Currently, the new SARS-CoV-2 and a genetically modified SARS-CoV-2
with a high fatality rate >83% identified during December 2019 have killed about
4.8 crore people worldwide. It is hard time for every science researcher to develop a
smart vaccine without side effects or a molecular efficient drug in a short time to
avoid the current uncontrollable infection. The standard protocols in medicinal
research involve defining the exact problem developing or creating a site and
design a lead drug or vaccine for genome infections. This methodology will require
long periods of innovative work, which requests colossal theory, which cannot
execute for the permanent virulent disease. Maybe we need a quick and worthwhile
come-up to hinder the viral disease. In silico strategies are a preexamination loom
before wet research facility studies, which enjoys the benefit of having the option to
make a quick and cost-productive investigation. Also, by selecting suitable algo-
rithms, we can exactly predict the reactive site on the virus from this; we can
Machine-based drug design to inhibit SARS-CoV-2 virus 297
elucidate the molecular structure of the vaccine candidate and a growth inhibition
drug for infected proteins. By choosing the target site, we can synthesize a lead
drug for the deadliest virus, which causes coronavirus-19 infection.
The scientists report speculation that inborn traces of coronavirus can incor-
porate into our cell chromosomes and stick there for a long time even after the
disease is cured. They tracked down those undeniable degrees of regular ribonu-
cleic acid (RNA) have created in light of the mix of infection with cell DNA that
peruses its grouping and makes a greater amount of it for duplication. It is
recommended that in uncommon discovering, individuals can recuperate from
coronavirus yet then test positive for coronavirus again months after the fact. The
scientists underlined that viral combination did not mean individuals who recup-
erated from coronavirus-19 stay irresistible. Since the COVID-19 antibodies
dependent on messenger RNA (mRNA) may be one way or another change human
DNA. The protein, switch transcriptase, is encoded by LINE-1 components,
groupings that litter 17% of the human set of chromosomes and address antiques of
earlier contaminations by retroviruses.
A genetic laboratory analysis reports that when human cells spike with addi-
tional LINE-1 components of coronavirus, DNA adaptation of SARS-CoV-2’s
groupings settle into the cell chromosomes. The experimental evidence from
“National Cancer Institute” supports the original hypothesis that the integrated viral
sequences will remain on human cells even after cured of coronavirus. The
advanced program that can be incorporated into in silico methods can assist
researchers in developing a drug that can target two or more diseases at one time
(combination therapy) [2]. In any case, the improvement of the best combos is the
most part difficult to make in a brief period span in a prudent and harmless to the
ecosystem. To smooth out combos dosing regimens, the improvement of particles
with double hindrances abilities against at least two unique classes of the objective
would be ideal [3]. It turns out to be more important, if clinically demonstrated
FDA-supported antivirus, virus tracking MRI contrast agents (CAs), monoclonal
antibody (mAb) drugs are considered for the studies. To develop smart multi-target
molecules for the current situation, the researcher needs to pick an appropriate
antiviral drug from the drug data bank that is already in use for other viral infec-
tions. After that, the researcher can study their limiting linkage with lung
TMPRSS2, angiotensin-converting enzyme 2 (ACE2), and RdRc (viral) proteins on
SARS-CoV-2.
This chapter will discuss in detail the computer-based designing program used
for the designing and pre-synthesis of an antiviral drug for COVID-19 infection.
The readers will gain extra knowledge about the process happening in COVID-19
virus replication and the in silico procedure adopted to identify the key amino acid,
which is the root cause for the viral replication and structural transmission. The
chapter will give in-depth knowledge about the status of antiviral drugs that are
under reinvestigation for their ability against COVID-19 infection. In addition, the
chapter throws limelight on the importance and application of MRI, dual imaging
MRI, and computerized tomography (CT) imaging modalities that are the best
among the best than PCR analysis.
The sequence of this chapter is in the following order. First, we have given a
detailed introduction based on a literature survey about the in silico methods used
in drug designing against COVID-19. Second, we discussed the types of COVID-
19 virus and their replication and stereochemical transforming mechanism, fol-
lowed by the role of antibody and vaccine available against COVID-19. Third, we
described the test available for COVID-19 infection. Fourth, we discussed the top-
notch in silico methods used in anti-COVID-19 drug research and their working
procedure. Fifth, we have given a summary of the previous four portions and the
future scope. Finally, we finished the chapter with conclusions followed by
references.
11.2 What is SARS-coronavirus-2?
SARS-CoV-2 quite fit with Coronaviridae, a family with huge enveloped RNA.
The RNA set of the chromosome of Coronaviridae extends up to 27–32 kb with full
of interlinked structural and nonstructural proteins.
The name SARS-CoV-2 is derived from the fact that approximately 82% of the
RNA set of the chromosome is identical to the SARS-coronavirus [4], 50% iden-
tical with MERS [5], and 79% resemblance with SARS. The X-ray diffracted 3D
crystal structure of SARS-CoV-2 has about 88% identical employing a set of cor-
onavirus identified in bats (bat-SLCoronavirus-ZC45 and bat-SLCoronavirus-
ZXC21). Coronaviruses are an infectious disease with thousands of thornlike active
proteins arranged in a throne pattern mostly on the boundary that could infect pigs,
bats, cows, birds, and other mammals. Furthermore, it mutates quickly to one
another and humans, causing diseases in the respiratory tract, lung, gastro-intestinal
(GI) tract, liver, kidney, and central nervous systems (Figure 11.1).
These infections are competent to be grouped into four typical varieties called
alpha COVID, beta COVID, MERS COVID, and SARS-COVID-2 [6]. In the set of
COVID-19 chromosomes, the 50 -replicase generates a polyprotein. A frameshift
occurs between ORF-1b and ORF-1a, resulting in the formation of two polypep-
tides: pp1ab and pp1a, which are subdivided into 16 non-underlying proteins (nsp-1
to nsp-10 and nsp-12 to nsp-16) whose mutual mutation is interceded by the main
protease (M-pro) (nsp-5, 3CL-pro (3CL protease)/M-pro) and nsp-3, PL-protease
SARS-CoV-2 (3D-Str)
(S1 and S2 spike proteins)
SARS-CoV-2
Figure 11.1 Microscopic and 3D molecular image of SARS-coronavirus-2

(PL-pro). The 30 -end modulates multiple core protein molecules that have epithelial
layer (M), spike (S) proteins, nucleocapsid (N), envelope (E), hemagglutinin-
esterase (HE), as well as eight movable proteins (p6, 3a, 3b, 7a, 7b, 8b, 9b, and
orf14). Approximately 29 highly active proteins are involved in SARS-CoV-2
transmission and replication in congregation cells [7]. Every structural protein
received the translational pathology from the viral subgenomic messenger called
vsgRNAs, which has the RNA cooperate. The crystal structure of RdRc on SARS-
CoV-2 revealed that one nsp-12 subunit (cooperate responsive catalytic subunit)
contains a nidovirus RdRc-associated nucleotidyltransferase (NiRAN) realm in the
nitrogen atom functioned array and anchoring part to connect the C-terminal RdRc
with NiRAN domain [8]. The catalytic site has ASP 760, ASP 761 (SDD), and SER
759 residues that lie in the pattern “C” used for the primer binding of RdRc with the
congregation. However, the M-pro looks like a calm enzyme but contributes
equally to RdRc in the regeneration of SARS-CoV-2. This enzyme consists of
about 305 asymmetric amino acid sequences through His41 and Cys145 catalytic
divectors and the substrate-binding mark positioned between the sandwiched
domains I and II. Tai et al. [9] put forward the truth that the amino acid present
between 331 and 524 residues of spike protein produces receptor-binding domain
(RBD) fragment that helps the SARS-CoV-2 to bind strongly with bat ACE2
(bACE2) and human ACE2 (hACE2) as receptors. Thus, this pointed protein sec-
tion is accountable for a mutual admission of SARS-CoV-2 and SARS-coronavirus
in human lung ACE2-state cells.
Every structural protein received the translational pathology from the viral
subgenomic messenger called vsgRNAs, which has the RNA cooperate (RdRc and
nsp-12) and the 30 -50 exo-RNA. The crystal structure of RdRc revealed that one nsp-
12 subunit contains a nidovirus RdRc-associated nucleotidyltransferase (NiRAN)
realm in the nitrogen atom functioned array and anchoring part to connect the C-
terminal RdRc with NiRAN domain [8]. The catalytic site has ASP 760, ASP 761
(SDD), and SER 759 residues that lie in the pattern “C” used for the primer binding
of RdRc with the congregation. However, the M-pro looks like a calm enzyme but
contributes equally to RdRc in the regeneration of SARS-CoV-2. This enzyme
consists of about 305 asymmetric amino acid sequences through His41 and Cys145
catalytic divectors and the substrate-binding mark positioned between the sand-
wiched domains I and II. Tai et al. [9] put forward the truth that the amino acid
present between 331 and 524 residues of spike protein produces RBD fragment that
helps the SARS-CoV-2 to bind strongly with bACE2 and hACE2 as receptors.
Thus, this pointed protein section is accountable for a mutual admission of SARS-
CoV-2 and SARS-coronavirus in human lung ACE2-state cells.
11.3 Mechanism of SARS-coronavirus-2 infection in

human
Coronavirus looks like spherical and has thousands of spike proteins at the per-
iphery. These spike proteins have many amino acid sequences through which they
bind over the protease domain congregation cell and undergo replication in
humans. Every human has transmembrane proteins, ACE2, in the lungs, GI track,
kidney, and brain. Initially, spike proteins attack the ACE2 enzyme through RBD
and undergo conformational changes by the fusion of viral molecular RNA with the
human congregation cell layer. After the dispersion interaction, the viral ribosome
goes into the congregation cell and creates more infections by replicating its
arrangement of chromosomes and causes viral contamination. Once tainted, the
viral subgenomic messenger, which has nsps, gets collected into multiethnic
cooperate composites, to establish a reaction appropriate for recording and repli-
cations of the viral arrangement of a chromosome. The viral chromosome sequence
on the site will develop virions with primary proteins [10].
11.4 How SARS-coronavirus-2 multiplies?
During the replication cycle inside the congregation, the RNA-subordinate RNA
cooperate (RdRc), nitrogen-functioned protein, long-chain ribosome, and modified
active ribosomal frameshifting locale of the infection are profoundly associated with
connection with RNA of the congregation cell. There are four primary useful desti-
nations installed in the entire ribosome: (a) the mRNA-restricting site; (b) the tRNA-
restricting locales that contain the aminoacyl spot, neutral peptide spot, and cleaving
function; (c) the peptidyl relocate center (PTP); (d) anchoring center for replication
[11]. It demonstrates that ACE2 is set to be a more essential enzyme to recombine
SARS-CoV-2 over RdRc. In the spreading stage, the high polar region combined
through RdRc, this repeated procedure supports the nonstop Virna RN acid replica-
tion. From molecular screening contemplates, it accounted for that the coarticulation
of angiotensin-converting and TMPRSS2 enzymes in the lung and stomach-related
tracks is a key for SARS-CoV-2 to enter the human tissue anatomy [12]. The report
additionally affirmed that amino acid groups like Hi-296, Ar-345, and Se-441 were
the reaction performing groups of TMPRSS2, which also contains the highly viral
replication supportive amino acids like Ar-435, Se-436, and Gly-464. To help the
assertion, the Ar-710 at ACE2 element from 697 to 716 assists the arrangement of the
ACE2–TMPRSS2 complex [13]. This connection supports the enzymatic change in
ACE2 receptors to make connections with the TMPRSS2 enzyme and Ar-652. This
site can be anchored by tracking the presence of the TMPRSS2 enzyme.
From the docking outcome (Figure 11.2), it is affirmed that the Arg652 is
situated at the suitable pose toward the protease catalectic group of three amino
acids. Also, Arg716 is available at the connection spot of ACE2, and Arg710 offers
more in making H-bonds to do appropriate protease-receptor cooperation [14]. The
M-pro protein comprises three spaces. A notable b-barrel is mostly taken after
chymotrypsin is found in spaces I (deposits 8–101) and II (buildups 102–184). Yet,
space III has remnants from 201 to 306 that contain a-helices. The missed remnant
from 185 to 200 is utilized to associate spaces II and III individually. It calls
attention to space III of M-pro. It is more fundamental for keeping up the proteo-
lytic modification, which occurs by holding the space II and the long circle remnant
TMPRSS2 protein (3D structure) ARG 697B

ARG 705B
ARG 708B
ASP 713B
GLU 638B
TYR 414C ARG 413C
ARG 710B
ARG 417C
ARG 652B
LYS 39DC GLY 464 C

GLU 382C
ASP 435C
SER 436C
A5P345C
HIS 296G
SER441C
Figure 11.2 The boundary of ACE2 enzyme in lung and TMPRSS2 proteins
185–200 in a chemically positive orientation or potentially situating the N-terminal

remnant that assumes a significant part for the reactant action of the enzyme. Any
medicament that ideally ties the anchor site of the M-pro pocket may restrain the
SARS-CoV-2’s replication.
As indicated by in silico crystallographic results, 26 key proteins from 29 other
proteins were anticipating as plausible antigens. Among these 26 key proteins,
spike “S” protein was chosen as the best immunization competitor as a result of
having a flag peptide, one transmembrane helix, negative GRAVY esteem, a major
subatomic weight, moderate aliphatic file, beta wrap themes just as having steady,
dissolvable, and non-hypersensitive highlights with a remarkable half-life period.
This element is a significant boundary that demonstrates that the protein can be
foreordained toward the secretory pathway. In addition, nsp-10, orf8, and orf7a
proteins with flag peptides were deliberate as probable antibody competitors. The
computer-designed antibody can have a polar binding functional group which can
buried in the curl-like protein sequence in the virus protein [15]. More contrast to
the conventional crystalline structure, the new sterically transformed apo archi-
tecture of its M-pro (PDB: 6Y2E) did lack an N3 ligand that is covalently con-
nected to a central protease in the old structure. The RMSD calculated against this
most recent M-pro 3D molecular structure with no binding ligand shows the lowest
mean rate, when a binding ligand is present. This research is crucial since the M-
pro’s pocket is inextricably linked to the edge of Glu166, implying that the salty
scaffold plays a significant role in the pocket’s eternal stability. The medication
must link with target invulnerable receptor molecules to produce a continual
resistant reaction. These receptors play an important role in detecting the saved
microorganism-related subatomic instances (PAMPs) on various organisms, similar
to infections, triggering natural resistant commencement, and coordination of the
diverse invulnerable comeback. The antibody or drug must link with these receptor
molecules to produce a continual resistant reaction. For such application, subatomic
level docking procedure will be adopted to find the binding capability of the drugs
with toll-like receptors (TLRs). The 3D structure of ‘TLRs’ collected from the PDB
bank will be subjected to Gaussian or AutoDock-based algorithms to optimize the

molecular energy. The optimized output image with low energy is used for mole-
cular docking endeavors. TLR2 and TLR4 were applied to recognize primary viral
proteins as a precursor to incendiary cytokine production. Likewise, a few inves-
tigations on SARS-coronavirus have published the size of TLR2 and TLR4 in the
creation of a successful invulnerable reaction [16].
11.5 Human antibody generation and role of vaccine

11.5.1 Immediate action of human antibody
The human immune system automatically generates antibodies like IgM, IgG, and
IgA against SARS-CoV-2 when it recognized the presence of a viral set of chro-
mosomes. Half of the affected tissues will receive the self-generated neutralizing
antibodies in 7 days. In 14 days, almost all the infected ones will receive the
antibody for them [17]. Initially, IgM antibody secretion level increases for the
opening week of SARS-coronavirus-2 infection; after consequent weeks it becomes
maxima and comes back to normal level in most healthy individuals [18,19]. We
can see the self-generated antibody IgG after 1 week of viral infection and it will
reach a higher level for 48 days and even until the reinfection happens. Another
self-generated antibody IgA will be seen from 4 to 10 days of infection. So the
presence of IgM, IgG, and IgA antibodies level will indicate the nature of the virus
that causes infection. Multiplex immunoassay study on SARS-CoV-2 antibody
confirms the presence of self-generated antibody in sera and saliva.
11.5.2 Role of synthetic vaccine on COVID-19

Natural medicine or computer-based synthesized chemicals could be given as a
vaccine for the current pandemic. Once the body gets vaccinated, it will be dis-
tinguished by the gathering’s inborn invulnerable framework through designing
acknowledgment receptors to sort the pathogen-associated molecular model,
PAMMs. This PAMM represented in vaccine antigen catches the attention of
neutrophils, monocytes, and dendritic cells, which are patrolling throughout the
body for 365 days. With the help of the pattern-recognition receptors, the con-
gregation cells will identify the probable danger and turn it into action when they
spot the pathogen. They change the appearance of the surface particles and aug-
ment favorable to incendiary cytokines. Chemokines bring about the extravasations
and fascination of granulocytes, monocytes, and habitual executioner T cells. This
results in creating a provocative aggravation microenvironment, in which mono-
ecious segregates into monocytes, and the immature dendritic unit enacts. It acti-
vates the growth of the dendrite unit, preceded by the activation of T and B
lymphocytes concerning the lymph nodes, by changing the expression of migration
receptors on the cell membrane.
Once the T cells get activated, they will distinguish into regulatory CD4þ
cells to sustain resistant tolerance. More and more of the immature dendritic cells
will migrate toward the lymph nodes when it identifies the protein vaccine
TH1 cytokine action sequence Action mechanism of a multi-targeting vaccine

Cytokines prime
Cytokine cytotoxic T cells for Vaccine
Vaccine Helper T
receptor action cell
MHC II
Cytotoxic TLR2/TLR2
T cell TLR4/MD2
mbrane
TH2 Plasma me
TH1 cytokine
cytokine
Epitope binding
causes cytotoxic Mal Tram MyD88 Mal
MHC I
ol TRIF
Epitopic T cells to divide Antigen Cytos MyD88
sequences and attack presenting cells IRAK4
recognized infected cells B cells are IRF 3
Helper T
by the MHC activated by
cell
cytokines IRAK1 IRAK2
Activated
plasma cell NF-NB
Few T cells Activated
Antigen converts into cytotoxic T Memory B cell
cell Nucleus
presenting cells memory cells
Acts on Inflammatory cytokines:

Antibodies
Memory T cell infected cells IL 1β, TNF-α, IL-8, IL-18
IFN responsive genes: IFNβ
TH2 cytokine action sequence
Infected cell
Figure 11.3 Interactions mechanism of TLR with multi-epitope vaccine
antiques. Simultaneously, management of antiques into less significant sections

occurs, resulting in the wrinkling of human leukocyte antiques (HLA) in human
MHC atoms on the cell surface. Class I major histocompatibility complex
(MHC)-clustered atoms make peptides from the antiques formed in the cytoske-
leton of debased cells realistic. The phagocyte antiques are fundamentally shown
on class II MHC particles [20]. CD4þ T cells respond to antigenic peptides dis-
played by MHC class II atomic groups, but CD8þ T cells respond to antigenic
peptides displayed by MHC class I peptide structures. Animated CD4þ T cells
release cytokines that cause an increase in B-cell production, which is required
for optimal antique formation (Figure 11.3).
11.6 Real-time COVID-19 identification test (RT-PCR)

Nuclear chain antiques and protein-based findings are used in current analytic
studies intended for the SARS-CoV-2 outbreak; nonetheless, popular nuclear chain
identification by rRT-PCR remains the highest quality level. In comparison to the
currently available serological tests, nucleic analyses have enhanced affectability
and explicitness for the viral site. The acknowledgment of SARS-CoV-2 in the
internal nose and throat depends upon rRT-PCR filling in as delicate, accurate, and
specific virus identification. The technique begins with sample collection and
segregation of viral RNA and its transformation into matching DNA. The matching
DNA gets enlarged by using Taq DNA collaboration. The rRT-PCR measures the
viral burden in 2 days, which has given in Figure 11.4.
Isothermal enhancement is applied to warm cycling-based nucleic corrosive
amplification [21]. Worked-on real-time PCR (RT-PCR) is currently accessible to
identify assorted locales of the SARS-CoV-2 arrangement of the chromosome [22].
It will identify the SARS-CoV-2’s RNA-subordinate S and RNA, which was
Stages from sample collection to COVID-19 infection conformation in RT-PCR analysis
Deactivated Purified
virus RNA
266 13,468 21,563 29,674

5' F1a F1b 3' Positive
RdRp
Fluorescence
E N
RT-qPCR~2 h per primer set
Purified RNA is reverse transcribed to cDNA
PCR amplification
Retrotranscription Threshold
Negative
Copies per reaction (Ct )
Figure 11.4 COVID-19 infection test procedure using rRT-PCR analyzer
participated in (RdRc)/helicase (Hel) proteins as well as in nucleocapsid (N). The

RdRc/Hel measurements are extremely sensitive approaches for detecting viruses.
11.6.1 Limitations of RT-PCR tool

Even though in silico technologies may examine large quantities of proteins for
vaccine formulation, researchers need several constraints to be aware of. In silico
investigations have acknowledged disadvantages, such as insufficient information
in databases, inadequacies in specialized software, including the use of inaccurate
inventory tools. As a result, it is critical to pick the ideal method of analysis and
experiment with different parameters to get the finest findings for in silico inves-
tigations. COVID-19 is now diagnosing using a combo of CT scans and rRT-PCR
data. Because RT-PCR examination is so common, it is crucial to look into the data
it provides to clinicians and policymakers. The requirement for experienced
laboratory personnel and long waited times for results may make nucleic acid
amplification more difficult. The global benchmark real-time quantitative PCR
(RT-qPCR) takes 4–6 h to complete, not adding the time it takes to transfer the
samples to the lab, which could take several days. Variation in viral RNA patterns
can alter actual RT-PCR results by targeting various viral genome regions.
Furthermore, due to viral evolution, falsified outcomes may occur [23]. Sample
storage, minimal nucleic acid filtration, and financial cost and waiting periods are
all disadvantages of RT-PCR tests [24]. The rate of PCR sample screening varies
depending on the sample type: stool (29%), nasopharyngeal scrub (63%), sputum
(72%–76%), bronchoalveolar lavage fluid (79–93%), and oropharyngeal scrub
(32%–48%).
Lack of specificity in immunodiagnostic techniques is associated with antigens
that are highly conserved to kill coronavirus with autoantibodies in autoimmune
conditions. Antibody-based methods are most effective in 7–11 days after exposure,
making them less useful for diagnosing an acute infection [25]. Although RT-qPCR
involves the current gold standard intended for nucleic acid recognition, novel
approaches such as pulse-controlled multiplication are being investigated. Pulse-
controlled multiplication requires no RNA excavation and can be completed in 10 min
using a tiny device. In the years ahead, S- and N-based antigen detection systems will
be used in conjunction with nucleic acid amplification tests to improve COVID-19
diagnostic accuracy at a low cost. Upcoming years, efforts to build new therapeutic
platforms could pay off if the analyses are exact, original, and uncomplicated to
execute, deliver results promptly, and are inexpensive to produce in mass [26].
11.7 Discussion on in silico methods in COVID-19 drug

research
In silico methods aid both scientists and physicians in developing molecular
structures for various protein sequences in SARS-CoV-2, locating a suitable target
on the virus surface to kill, and designing a drug suit for the anti-bio war against
this most lethal virus. It will provide a detailed antimicrobial summary of the new
or repurposed drug for other deadly diseases.
11.7.1 In silico-assisted anchoring site analysis

Several PC-assisted programs, such as Glide, AutoDock Vina, MOE DOCK,
ClusPro, GOLD, and rDOCK, have been developed in recent years for both cor-
porate and academic purposes to predict the inhibiting mechanism and segregation
of a ligand in comparison to a protein.
11.7.1.1 AutoDock Vina

An automatic docking tool is an expansion for “AutoDock.” It is a user-friendly
procedure available in the free version to predict how molecules bind to a known
3D receptor [27]. AutoDock is the most generally acknowledged atomic group
docking apparatus that is comprehensively utilized for the survey of mixtures lined
up with likely targets. It is used in multidisciplinary approaches like X-ray crys-
tallography, structure-based medicine designing, lead molecule generation, com-
binatorics topology archive planning, proteinase-proteinaceous docking, virtual
screening (high-throughput screening [HTS]), and synthetic system consideration.
Right now, the AutoDock, 4.2 renditions are utilized for a molecular anchor
examination of ACE2, TMPRSS2, and RdRc protein against every one of the early
announced antiviral physic. AutoDock 4.2 is quicker than AutoDock, and it grants
side chains in the macromolecule to be flexible. The functioning proficiency is
more than the previous rendition, and firm anchoring is blindingly expedient, and
premium quality adaptable anchoring should be possible in 60 s. Of course, up to
40,000 unbending anchoring can be made in 24 h on one CPU. The improved 4.2
rendition of AutoDock has a free-energy scoring task dependent on a direct relapse
examination, the AMBER power field, and a significantly greater arrangement of
changed protein–ligand buildings with known save constants that could be recog-
nized. The best model was consistently cross-approved with a different
arrangement of human immunodeficiency virus (HIV)-1 protease buildings and

affirmed that the standard deviation from the original value was lesser than 2.5
kcal/mol. This is an adequate sum to separate between leads with nanomolar,
miniature, and milli-hindrance constants. Here some of the new highlights from
AutoDock 4.0 are added to the 4.2 renditions. The process gives the Monte Carlo
reproduced strengthening (SA) strategy for 2.4 and prior. The Lamarckian genetic
algorithm is a major enhancement for the genetic algorithm, and both hereditary
techniques are considerably more productive and powerful than SA. For its infor-
mation and yield, Vina utilizes a similar PDBQT subatomic construction document
design via AutoDock. PDBQT records can be created from Open Babel GUI pro-
gramming and observed by using MGL tools. The yield records with restricting
energy esteem can be seen through the Python program.
11.7.1.2 Molecular dynamics

It is a method for splitting down the correct activities of particles and molecules on
a computer. The host molecule and the newly designed antiviral drug molecules are
allowed to link with one another for a set amount of time, providing insight into the
application’s dynamic “growth” [28]. It is a Python library to dissect directions
from molecule to atom elements (MD) recreations in numerous famous organiza-
tions. It can compose the majority of these arrangements, as well, along with iota
determinations reasonable for representation or local investigation devices. This
analytical procedure permits anyone to peruse molecule-based directions and
access the nuclear is arranged through NumPy exhibits. It gives an adaptable and
generally quick system for complex investigation assignments. Directions can
likewise be controlled and worked out. The MD works on well-known simulation
programs like NAMD-10, Amber, Gromacs, and with a series of preparation and
analysis tools, joined together in a common interface. To discover the commu-
nications of physic with anchoring site Newton’s condition of movement shall be
applied to this MD framework until the molecules are allowed to connect for a
while at a given temperature that observes the law of traditional mechanics. Also,
in a constant test, the proportion of trial upsides of tests is performed through
estimating instruments regarding time. In the wake of playing out the harmony of
the framework, direction investigation and estimation is done, which is equivalent
to the examination of the example estimations like factual investigation of esti-
mated perusing. The complete protein structure will be obtained by dissolving it in
a water box with parameters LWH of 97.42 Å in the physic evaluation. The addi-
tion of neutralizing counter ions like Kþ and Cl to the protein molecular structure
nullifies the protein surface charges and it is tuned roughly to 0.15 M. The van der
Waals (vdW) energy can then be calculated using a smooth cutoff (10–12 Å).
11.7.1.3 The DockThor-VS platform

The DockThor docking program was engineered to administer fundamentally 1D
and 2D ligands with zero steric factors that have excellent peptide binding or
docking efficiency than the other regular branched ligands with restricted move-
ment. The running program utilizes a phenotypic gathering subject to different
strategy solid-state hereditary calculations as solicitation method [29]. Here, the

parental substitution system works based on dynamic modified restricted match
selection. It provides an unrivaled evaluation that awards perceiving diverse simple
planning in a single run by assessing the hyper-surface energy and obtaining
everyone grouping of the conveyed structures. These scoring cutoff points will take
the going with big data terms for protein–ligand limiting: (a) a curved continuous
disorder term that repels the “steric or static” rotatable protections, (b) inter-
molecular affiliations terms, (c) an ionic active site that has adequate polar joint
initiating the process of binding togetherness of the newly designed ligand over the
anchoring site on the M-pro, (d) a positive no polar solvation term that is relative
just to the dissolvable open surface, and (e) a protein–ligand lipophilic
correspondence term.
As far as possible, it will be prepared and endeavored in an immense set
(>2,900) of top-of-the-line 3D improvements related to gathered physicochemical
profiles and huge recuperating fixations for drug plans. The docking algorithm
works on the request of the definite file format, usually a geography doc. will be
preferred for the drug molecule, for cofactors, top format will be preferred, and for
recording any protein of our interest, in form containing the iota types and inade-
quate charges from the MMFF94S49 power field will be the required format. The
entire atomic power field limits performed by implying assignments without the
clients’ need intercession. The pursuit space tended to as an association box where
the potential results are dealt with at the construction communities, fundamentally
lessening the computational expense. The hidden individual group is abstractly
conveyed inside the lattice box utilizing erratic qualities for the rotational, trans-
lational, and conformational levels of the possibility of the ligand. For every SARS-
CoV-2 strong objective, DockThor-VS platform gives a proposed put-down of
places to pause for the association box that the client can utilize or change as
demonstrated by the protests of his docking test.
11.7.2 Machine-assisted designing and evaluation of

COVID-19 drug
Computational screening is currently becoming the primary emphasis for resol-
ving the SARS-CoV-2 infection outbreak. Silicon-based intelligent processors
with powerful viral evaluation algorithms have emerged as dominating tools in
the pharmaceutical industry and have been utilized in vitro [30] for more than a
couple of decades to identify protein blockers and understand protein–protein and
protein interactions. Although millions of antiviral medications are available
online, numerous experimental, clinical, and computational drug research efforts
have been launched to quickly prioritize compounds to recognize them as
potential therapeutic drugs for the SARS-CoV-2 disease outbreak. This strategy
discloses lucid pathways for the researchers to design and develop similar
effective smart drugs in less time. To restudy the efficiency of the drug for the
current situation, the researchers can use the data available at the antiviral drug
library. From Selleckchem Inc., Kyoto Encyclopedia of Genes, Genomes, and
DrugBank databases, the researcher can retrieve the US-approved drug library
and Europe-based FDA drugs. Reinvestigating existing antiviral drugs as a

potential drug candidate will assist scientists in exploring a proposal for medi-
cines that have well-established pharmacological and safety profiles [31].
Chloroquine and hydroxychloroquine are the two evergreen antiviral drugs that
are used for several decades to treat a variety of diseases that have been thor-
oughly researched and approved for long-term use. Antiviral drug reevaluation
using a computational program could be undertaken at any stage of pharmaceu-
tical research along with any number of drug sets (Chen et al. [32] used about
7,173 for screening). The target protein and its matching 3D crystalline structure
of protein databases such as PDBe, PDBj, RCSB, and BMRB will be the next and
perhaps most crucial steps. After selecting the 3D structure using platforms like
molecular docking, molecular mechanics calculations, and a combination of
quantum mechanics, the investigator can uncover a wide range of active com-
pounds and spot possible molecular mechanisms for the target sites of these well-
practiced pharmacologic agents [33]. To do molecular dynamics simulations on
the protein targets, the researcher should select the drugs through scoring meth-
ods, followed by validations [34]. The following are the most trusted methodol-
ogies that are in use to find the protein target and to find the exact antiviral drug
for COVID-19 treatment.
11.7.2.1 Molecular docking program

Molecular docking and chemical binding techniques were used to evaluate the
pharmaceutical efficacy of antiviral medicines and evaluate them toward enzyme
systems of the SARS-CoV-2.
For SARS-CoV-2 treatment, most of the binding studies focus on vital pro-
teases like 3CLpro, S spike, E envelope, methyltransferase proteins, and RNA
polymerase [35]. To determine the affinities over 3CLpro, by molecular modeling,
the previously approved drug molecules were subjected to molecular docking
study. The study shows that lopinavir, nelfinavir, ritonavir drug are best binding at
23.8%, remdesivir binds at 19.0%, and darunavir binds at 14.3% of the total
number of the drug studied (Table 11.1). Also, the author insists that the majority of
the drugs show no binding over the SARS-CoV-2 virus.
Based on the results received, most of the drugs are now in clinical trials. The
19 virus–transcribed proteins are screened by Wu et al. [36]. The drugs were fore-
seeing for specific proteins, and for chloroquine drugs, the 3CLpro and PLpro were
fixed as a target. And for remdesivir drug, proteases like 3CLpro and RdRc have
been fixed. The docking simulations show that remdesivir displays better bind
affinity than chloroquine drug over the 3CL-pro M-pro. Another study used
AutoDock Vina to evaluate 19 drugs against the M-pro site of SARS-CoV-2. The
drugs are O6K, chloroquine, hydroxychloroquine, remdesivir, favipiravir, umifeno-
vir, bromhexine, ambroxol, montelukast, dipyridamole, nelfinavir, cinaserin, GS-
441524, curcumin, quercetin, kaempferol, lopinavir, entecavir, and N3. O6K and
N3 have outstanding binding affinity values of 7.12 and 7.40 kcal/mol, corre-
spondingly. The docking images show that the molecule has the best docking scores
and hence anchors the M-pro’s niche throughout the crystalline structure. Quinoline
Table 11.1 List of approved natural and synthetic drugs that are in trials against
COVID-19 outbreak
S. no. Targeted proteins Antiviral and other Studies in

under study chemical drugs progress
1 3CL-pro Lopinavir 9
2 3CL-pro Ritonavir 9
3 3CL-pro Nelfinavir 7
4 3CL-pro Remdesivir 9
5 3CL-pro Darunavir 6
6 3CL-pro Atazanavir 5
7 3CL-pro Ledipasvir 5
8 3CL-pro Oseltamivir 5
9 RdRc Galidesivir 4
10 RdRc Remdesivir 6
11 RdRc Ribavirin 4
12 3CL-pro Captopril 4
13 3CL-pro Chloroquine 6
14 3CL-pro Etoposide 4
15 3CL-pro Hesperidin 3
16 3CL-pro Hydroxychloroquine 6
17 3CL-pro Imatinib 2
18 PL-pro Chloroquine 3
19 PL-pro Darunavir 2
20 PL-pro Levamisole 2
21 PL-pro Lopinavir 4
22 PL-pro Ritonavir 4
23 RdRc Favipiravir 5
24 RdRc Sofosbuvir 2
25 RdRc Tenofovir 2
26 “S” spike þACE complex interface Tazobactam 2
N-oxide (QO) and Quinoline (Q) nitro compounds were analytically explored
(Figure 11.5).
The medication candidates are associated well with the M-pro active site, with
binding energies ranging from 4.31 to 5.04 kcal/mol. In addition, some of the
nitro-QO compound inhibitors, such as N9-QO, N8-QO, and N4-QO, showed
slightly higher stabilizing binding energy values than their nitro-quinoline coun-
terparts. Docking tests were done with the commercialized medications CQ and
HCQ, which are presently used to treat SARS-CoV-2 contamination, to test the
possibility of such discoveries, in addition to their binding, energy conditions were
determined to be 2.33 and 2.80 kcal/mol, correspondingly (Table 11.2).
The hydrogen-bonded relation of the N4-Q molecule with Cys44 at 2.78 Å and
hydrophobic interactions with other amino acid sequences like Cys145, Gln189,
Glu166, Gly143, His41, His163, His164, Met49, Met165, Ser46, Ser144, Thr45,
and Val42, respectively, are given in Figure 11.6 (MD-1). These linkages are
D-1 D-2 D-3
C145
H41
H41 C145
E166
E166
HOH
HOH H41
HOH
D-4 D-5 D-6

E166
E166
H164
C145
C145
H41 HOH
HOH H41
E166
HOH
Figure 11.5 Binding structure of (D-1) N4-Q, (D-2) N4-QO, (D-3) CQ, (D-4)
CQO, (D-5) HCQ, and (D-6) HCQO in the M-pro active site
Table 11.2 Binding sites and their affinity energy for the antiviral drugs
chloroquine and hydroxychloroquine
CH3 BS-3
BS-2 CH3 CH3 OH
H N CH3 H N CH3
N N
BS-3 BS-2
Cl N Cl N
BS-1 BS-1
Protein-binding Affinity energy in Protein-binding Affinity energy in

sites (BS) kcal/mol sites (BS) kcal/mol
BS 1,2 2.93 BS 1,2 3.22
BS 1,3 2.42 BS 1,3 2.93
BS 1,2,3 2.61 BS 1,2,3 2.84
BS 2 3.03 BS 2 3.45
BS 2,3 3.02 BS 2,3 3.34
BS 3 2.94 BS 3 3.31
MET
THR MD-1 MD-2 A:165 PRO
A:45 A:39
GLN HIS
CYS A:189 A:164
A:44
GLU
HIS
A:166
A:41
SER GLY SER
A:46 A:143 A:46
MET
A:49 GLN SER
HIS
A:189 A:144
A:163
CYS
MET A:145
HIS A:49 SER
A:41 VAL CYS
MET GLU HIS A:147
A:42 A:145 GLY
A:165 A:166 A:163 A:146
LEU
A:27
HIS GLY SER
A:164 A:143 A:144
Figure 11.6 Image of molecular dynamics analysis done for N4-Q and N4-QO
with M-pro
necessary for decreasing M-pro’s enzyme reactions, and they are consistent with
the findings of earlier docking studies [37]. The N4-QO molecule, on the other
hand, binds to the M-pro was stabilized by four H-bonds such as Ser144 (3.04 Å),
Gly143 (3.01 Å), His41 (2.93 Å), and His163 (2.72 Å), and hydrophobic affinity
with Cys145, Glu166, Gly146, Gln189, His164, Met49, Met165, Leu27, Pro39,
Ser46, and Ser147, respectively, Figure 11.6 (MD-2). Interestingly, the binding
affinity of HCQO gets raised to 1 kcal/mol, which confirms their stability than any
other derivatives of CQ. Possible variations of HCQO and CQO compounds could
be used for evaluating the weight of the N-oxide group at different sites. Every
medication nominee’s molecular modeling position suggested that they can anchor
precisely within the membrane chamber. According to molecular docking and
binding energy calculations, viral proteases that are responsible for hewing the viral
polypeptides and maintaining the viral regeneration process could be ideal ther-
apeutic targets for viral illnesses. In cases where the virus generated confrontation
by transmutation, protease inhibitors have been employed in combination medi-
cation therapy. Since the M-pro enzyme contains two polyproteins, pp1a and pp1ab
act as the most important foundation for virus duplication. HTVS, SP, and XP-
based gliding scores are used to create hit molecules using a docking-based mole-
cular modeling approach, while ADMET can then be used to depict the pharma-
cologic and physicochemical properties of the suggested lead compounds.
To investigate the affinities of the lead compounds, molecular concept mod-
eling might be used. When comparing the binding affinity of the molecule ABBV-
744 to M-pro (G-bind 45.43 kcal/mol), it is discovered that the combination is more
stable than other protein–ligand complexes. It is a hypothesis that remdesivir
attaches to targets, RdRc and 3CLpro on SARS-CoV-2 with high interaction score.
According to the results of a molecular docking study, remdesivir and its enzyme
(GS-441524) have excellent binding scores of 6.4 and 7.0 kcal/mol, respec-
tively, and stability that with M-pro (Figure 11.7), indicating supplemental
–5
Affinity (kcal/mol)
–6
–7
te ine
F he e
Di avip xine
yc b le
M oro xol
GS nas st
Ka -441 rin
4
En ina l
d n
in
N3
ox Am amo r
Um tec vir
Qu nov r
Re erc ir
Cu fina ir
rc vir
K
p o
id vi
ife avi
om uin
em 52
m eti
l v
Ci luka
O6
Lo pfer
um
on qu
e
yr ira
Ne esi
hl ro
Br roq
lo
Ch
p
dr
Hy
Figure 11.7 Graphical image showing anchoring efficiency of various antiviral

drugs in the main protease-anchoring site
stimulatory activity in furthermore to its RdRc inhibition impacts. Remdesivir’s

binding index is one of the highest therapies, making it an excellent choice
designed for the initial treatment of COVID-19 patients for antiviral therapy.
Favipiravir, a tiny viral growth restrictor compound for RdRc produced by
Fujifilm Corp. in Japan, did not demonstrate notable interaction with the primary
proteolytic. Entecavir has a similar molecular chain just like the remdesivir’s
metabolite (GS-441524) that is abundantly available at a low price, which can be a
better alternative as a potential RdRc and primary target proteins. The properly
suited interaction value for entecavir, according to docking study data, is 6.41
kcal/mol. Murugan et al. [38] used structure-based quantitative diagnostic criteria
(SBCSA) on the drugs such as dihydroergotamine, phthalocyanine, lonafarnib,
nilotinib, R-428, tadalafil, and baloxavir marboxil extracted from DBD to evaluate
molecules against some of the most prevalent infectious targets, such as 3CL-pro,
PL-pro, RdRc, and the spike protein. For the prevention of the COVID-19 virus,
baloxavir marboxil is now undergoing drug trials. They claim that drugs like
dihydroergotamine, phthalocyanine, R-428, nilotinib, tadalafil, lonafarnib can bind
to all three targets in a single attempt making them multi-targeting medicines.
Curcumin has a low absorption (1%) that could limit its use in treating persis-
tent viral illness. Curcumin taken orally, on the other hand, may accumulate in large
concentrations in the liver and GI system. It protects the GI from the SARS-CoV-2
virus. It is proved that curcumin is an effective medicine for preventing viral
infections in the GI system during medication administration and can continue to
play a role even after the respiratory infection has spread to the lungs. Besides these
critical functions, this instinctual drug has an adaptable feature in curing swelling,
including IL-6, IL-1b, and TNF-a. Also known for preventing liver and digestive
tract or GI tract injury, it would be considered for wide-spreading COVID-19
pathological circumstance and was also widely acknowledged to prevent liver and
GI tract rupture, which causes diarrhea. Curcumin stabilizes the interaction with
SARS-CoV-2 major protease, according to molecular modeling experiments, with
better interaction energy of 7.1 kcal/mol. When it creates stable compounds
with the M-pro, the other antimicrobial drugs quercetin and kaempferol, which
seem to be abundant in fruits and vegetables, have greater docking studies values
of 6.45 and 6.62 kcal/mol, correspondingly (Figure 11.7).
Nelfinavir is used to treat HIV, and a molecular modeling investigation of such
an antiviral drug revealed a kind of approaches, a score of 7.01 kcal/mol with the
SARS-CoV-2, major proteases. The higher bonding strength indicates their ther-
apeutic implications for inhibiting M-pro in COVID-19 illness. It is currently being
tested in clinical trials as an antiretroviral to prevent SARS-CoV-2 replication.
Umifenovir is an antimicrobial agent available in Russia and China that has been
experimentally used in China to manage COVID-19. In vitro, umifenovir reduces
SARS-CoV-2 with an IC50 of 30 M [39]. It has also demonstrated significant
benefits as COVID-19 distributed postexposure prophylaxis. According to a
molecular modeling study, the M-pro site of SARS-CoV-2 can engage with the
major enzyme and limit replicating with an interactive score of 6.51 kcal/mol. An
asthma inhibitor molecular drug, montelukast, shows excellent suppression of M-
pro, as evidenced by the interaction affinity 6.2 kcal/mol acquired via docking
studies. Montelukast has also been shown to be effective as an anti-inflammatory
agent in the SARS-CoV-2. Another medication drug dipyridamole has a wide range
of antiviral properties and has been clinically demonstrated to be an excellent anti-
COVID-19 drug [40]. Dipyridamole’s antiviral mechanism includes the inhibition
of the M-pro, as demonstrated by an in vitro surface plasmon resonance experi-
ment. The binding energy score is 5.8 kcal/mol.
11.7.2.2 High ambiguity–driven protein–protein DOCKing

Protein–proteins, protein–nucleic acid, and protein–ligand complexes are among
the many types of modeling challenges it can handle. High ambiguity–driven
protein–protein DOCKing (HADDOCK) found 80 frameworks in 11 tools,
including the receptors-2 (TLR2) clusters (s), accounting for 40.0% of the water
refined HADDOCK models [41]. The negative HADDOCK score of
112.0 2.850 indicated the vaccine’s high binding affinity with the receptors. On
human TLR2 receptors, the CTL-1 and CTL-5 epitopes are bounded on SARS-CoV-
2 infection. TLR2’s interactions with the legend activate the immune system’s
supervisor and cause interferon and interleukin to combat the infection. According to
the well-known Ramachandran design inspection, 68.9% of TLR2 structural residues
are found in the positive zone, 31.13% in the allowed zone, and none in the outlier
region. The binding results revealed that CTL-5 (WTA GAA AYY) connected pos-
teriorly with the TLR2 receptor’s docking groove. The binding score reflects roughly
16.5% of the water-refined HADDOCK generated models. HADDOCK program
clustered 33 structures in 7 TLR4 cluster(s) and shows negative gain. The
HADDOCK score of 130.9 10.1 indicates good binding energy between the
vaccine and the receptor. Here the vaccine masked nearly 2,204.4 22.4 surface area
on the target protein which is less exposed to water (hydrophobic protein surface).
The docked complex with low RMSD score (Table 11.3) implies a high-
quality model. HADDOCK grouped 120 structures into 12 MHC class I clusters,
accounting for 60.0% of the water-sophisticated HADDOCK mock-up. The low
HADDOCK score of 214.74.1 shows that the antibody and MHC class I
receptor has significant affinities, and the reduced RMSD indicates that the
docked complex is stable. HADDOCK groups 64 structures into nine MHC class
II clusters, accounting for 32% of the water-refined HADDOCK stimulate
models. The stable and strong molecular interaction between the virus active
spot and vaccine is proved by the low HADDOCK score. The human MHC class
I cell surface receptor HLA-A*0201 is one of the most common ones. HLA-
A*0201 has a three-dimensional shape found in the RCSB protein database
under the PDB ID 1I4F (Figure 11.8). Chain-A with 275 amino acids and the
beta-2 microglobulin chain, which has 100 amino acids, make up HLA-A*0201
[42].
Table 11.3 HADDOCK binding values for TLR-2, TLR-4, MHC-I, and MHC-II
with M-protease
No. Docking parameters TLR-2 TLR-4 MHC-I MHC-II

1 HADDOCK value (a.u.) 112.0 130.9 214.7 212.1
2 Cluster size 20 20 20 20
3 RMSD (Å) 0.3 0.3 0.3 0.3
4 Van der Waals energy (kcal/mol) 73.2 72.4 138.5 132.5
5 Electrostatic energy (kcal/mol) 319.7 238.9 156.3 394.9
6 Desolvation energy (kcal/mol) 25.1 10.9 45 0.6
7 Restraints violation energy (kcal/mol) 0.01 1.1 0.0 0.2
8 Buried surface area (Å2) 2,094.7 2,204.4 3,585.9 4,276.9
Figure 11.8 A 3D structure of HLA-A*0201 receptor (PDB ID: 1I4F)

180
~b
b
135 b ~b
~I
90
I
45
a
0
~a
–45
–90
ASP 29 (A)
–135 b ~p
p
b
~b
–180 –135 –90 –45 0 45 90 135 180
Figure 11.9 The image showing the Phi (degree) vs Psi (degree) for HLA-A*0201
(Ramachandran plot)
The Ramachandran plot is adopted to examine the HLA-A*0201 structure for

stereochemical characteristics. It shows an increase that approximately 93.72% of
the receptor residues are in good agreement, 6.3% of the receptors are all in
strong agreement that cannot be vomited from consideration, and therefore most
importantly, no antibody residue seems to be in the out-of-limit region
(Figure 11.9).
Furthermore, verification of a 3D server demonstrated high structural quality,
with over 97.1% of proteins having average 3D–1D scores of 0.2. The HPEPDock
server is adopted to analyze the binding of chosen epitopes using the optimized
HLA-A*0201 structure. The PepFold administration is adopted to control the 3D
designs of chosen epitopes. The energy of planned 3D constructions of CTL
epitopes was diminished and upgraded for subatomic docking concentrates with
the HLA-A*0201 got. As indicated by the information, CTL-1 and CTL-5 epi-
topes show good adsorption quantum scores of 251.779 and 246.834 kcal/mol,
separately.
11.7.2.3 Molecular dynamics

Though we have advanced techniques for the identification of suitable drugs against
COVID-19 infections, we should not forget the chance of getting instrumental and
human error during report generation. The combination of computational algorithms
like Molecular Dynamics (MD) and bindery free energy assessment on any drug
molecules will be a wise choice for distinguishing the potential antidrug capacity
among them [43–45]. The MD reenactments and conformational examination
anticipated the steady connections of grazoprevir with ACE2, TMPRSS2, and RdRc
proteins to make them the most promising mediation for COVID-19. However, these
collaborations are noncovalent and transient. vdW force is a major commitment in
confining free energy for all the structures that make them more stable and steady.
These values can be calculated from the atomic mechanics-Poisson Boltzmann sur-
face area examination. An in silico approach is used for upgrading the limiting
between a planned immunizer mAb C-B6 and RBD of S protein. The MD repro-
ductions, just as FEP computations, give way to various cheerful contenders for
upgrading the C-B6-DAW and C-B6-AME antibodies. Despite being expected, they
distinguished a few key RBD deposits, specifically K-417, L-455, N-487, K-458, S-
477, and G-476, which were significant for RBDFAB cooperation. It is worth
noting that among these microstructures, K-417, L-455, and N-487 are found to be
more suited for RBDACE2 connections [46]. This investigation affirms the hin-
drance capacity of the monoclonal neutralizer against the COVID. Today, killing
mAbs had been isolated from platelets of recently recovered COVID-19 patients and
used as an antiviral specialist against the S spike glycoprotein as a day-by-day life
strong treatment for coronavirus-influenced patients. Since this is an unsafe metho-
dology, the specialists adjusted the methodology with known mAb with S-protein.
Using the in silico strategy, the scientist examines the atomic system of mAb’s
limiting with S-protein and plans more powerful mAbs through protein mutagenesis
considers. For MD examination of independent apo M-pro just as the ligand-bound
particle, the IBM Power-bunch fused with NAMD-10 adaptation was performed. The
TIP-3P model for water, the standard particle power field, and the CHARMM-36
power field for the protein are adopted. Before executing the exam, each molecule in
the M-pro in the SARS-CoV-2 3D crystal structure in the biochemical and physio-
logical environment was homogenized.
Molecular dimensional analysis is adopted to find the structural stability of
the molecule when it binds to the target protein at least for 100 ns. As given in the
previous discussion, the repurposing of mitoxantrone and leucovorin drug from
the FDA library is adopted for analysis against COVID-19 treatment. Likewise,
two more antiviral drugs, birinapant and dynasore, were selected for analysis
against the SARS-CoV-2 M-pro site. The chemical structure of the molecules in
the antiviral molecule and their steric energy study for a minimum of 100 ns
adopted on these antiviral drugs shows an RMSD value of 2.25 Å. These unex-
pected superior values are due to the improved steric and hydrophobic nature
surrounded in the compact cavity when the drug binds over the viral protein.
Glutamic acid head (Glu-166) of the principle protease is a critical buildup to
hold and frame a steady unpredictablity of revealed lead compounds by shaping

hydrogen bonds and salt scaffolds. Additionally, the hydrogen bonding prob-
ability of MM/GBSA and its free energy computations confirm the steady and
firm binding of the drug molecules with SARS-CoV-2 M-pro protein.
Lycorine, a characteristic alkaloid, accounted for to have various organic prop-
erties, like an anti-infection, against the tumor, hostile to the parasite, and bacteria.
Emetine, a natural drug molecule has capacities, including prophylactic exercises,
and hostile to infection, and against malignancy. The counter SARS-CoV-2 exercises
of lycorine and emetine report a similar result. Lycorine destroy SARS-CoV-2 with
EC50 ¼ 0.312 mM and emetine with EC50 ¼ 0.463 mM [47]. The revealed hostile to
infection movement of lycorine resembles our own (EC50 ¼ 0.437 mM). The com-
putational recreations demonstrate that these decidedly charged mixtures could
capably join with mono negative RPF of SARS-CoV-2, adding to weakening the
scattering of the infection. Furthermore, it accounts for that the SARS-CoV-2 infec-
tion can spread its disease on the cerebrum tissues [48]. Lycorine is less harmful and
can cross the blood–cerebrum boundary and fix brain contamination. Vinifera,
nympholide A, myricitrin, hesperidin, afzelin, phyllaemblicin B, and biorobin are
naturally occurring antiviral and antibiotic drug molecules with high binding affinity
on the main proteins and amino acids with nitrogen function proteins on SARA-
coronavirus-2. But antiviral drug treatment over ACE2 and RdRc protein target are
suppose to have less effect when compare to other targets. So more focused and
target cum sequence specific antidrug binding practice is yet to perform and each and
every proof against these targets should be noted with utmost care.
Few medications that can hinder the replication cycle of the infection have been
proposed and tried with atomic docking that concentrates as a component of the most
recent clinical treatment draws near. These medications incorporate chloroquine and
hydroxychloroquine. Hydroxychloroquine is more dissolvable than chloroquine and
delivers a lesser measure of destructive metabolites. Although chloroquine and
hydroxychloroquine, which have been in clinical practice throughout the previous 70
years, are drugs for the treatment of immune system illness, they are utilized simi-
larly as antimalarial drugs. These mixtures have been accounted for as potential
expansive range antiviral medications for COVID-19; in any case, their advantages
against COVID-19 are questionable, with no obvious impact on hospitalized patients.
From an atomic perspective, these drugs are thought to inhibit viral transformation.
While increasing the endosomal pH, the drug infection, cell fusion, interfered with
the glycosylation of SARS-coronavirus cell receptors, are raised. They also play an
important role in the resistance adjusting process, which has a synergistic antiviral
effect in vivo. The QM/MM-enhanced molecular dynamics study on the main anti-
viral drug remdesivir in its monophosphate is adopted to discover the chance of
joining with the preliminary strand mRNA connecting with RdRc (PDB; 7BV2)
molecule as given in Figure 11.10.
The molecular surface binding investigation of remdesivir mono-phosphate
over RdRc shows striking communications that incorporate alluring vdW, pp
stacking, p-cation, p-alkyl, and covalent just as appalling electrostatic collabora-
tions with RdRc practical gatherings.
T:U10
P:U20
Figure 11.10 QM/MM output image showing the anchoring of remdesivir on

helical mRNA genome present in RdRc protease
11.8 Machine-integrated advanced techniques for

COVID-19
From the in silico algorithm, the scientists can acquire in-depth knowledge about
the newly designed or existing drug properties like binding efficiency on M-pro,
antimicrobial property, virus replications inhibition efficiency, and multi-targeting
nature. The challenge is to study the in vivo effect of drugs on humans. For this, the
most promising techniques like CT and MRI scans are adopted to find the viral
infection and the curing ratio of the disease. Nowadays, due to the research in
machine learning methods, AI, IoT, etc., these imaging techniques are adopted to
get more promising results than the classical ones. Almost every physician, irre-
spective of their specializations, is dependent on these techniques and based on the
computerized output, they are deciding the treatment protocols.
11.8.1 Computerized tomography in COVID-19 detection

To create cross-sectional images of the body, a CT screening uses computers and
rotating X-beam machines. Such image provides more information than standard X-
beam images. A CT sweep may give 3D pictures of the inner organs just like the
clinician envisioning the head, delicate tissues, veins, bones, shoulders, spine, heart,
midsection, knee, and chest. During a CT analysis, the machine turns and takes an
arrangement of X-beams from different points. The PC-based program will assist us
with consolidating the posture to make cross-segments pictures of the body. The
region of interest (ROI) on any human internal organ is collected as a 3D picture
under CT. A CT check typically exhibits a “switched radiance” model and crypto-
gram of septal thickening [49]. COVID-19 CT images clearly show the corre-
sponding pneumonic parenchymal surface along with integrated pneumonic opacities
having flattened morphology and little lung dissipation. Both MERS-coronavirus and
Ground-glass
opacification
Normal SARS-CoV-2
infection
Figure 11.11 Coronavirus-infected lung image obtained on CT scan
Tissue
damage
Normal SARS-CoV-2
infection
Figure 11.12 Coronavirus-infected brain image obtained on MRI scan
SARS-coronavirus patients have lung fixation with periphery predominance [50].

People with negative rRT-PCR test results for SARS-CoV-2 can have chest infection
when tested under CT scan and without further determination to have COVID-19.
Subsequently, rising evidence ropes the utilization of chest imaging assessment
mostly as positive intended for COVID-19 sickness when people contain negative
rRT-PCR tests. Various investigations have examined chest pictures of people tainted
by SARS-CoV-2, taking into account that irregularities may likewise be because of
different reasons of pneumonia, prompting bogus positive outcomes (Figure 11.11).
In cutting-edge cases, SARS-CoV-2 disease can cause lung tissue harmed with
diminished oxygen take-up in contaminated people. The insusceptible reaction to
SARS-CoV-2 prompts the release of cytokines and chemokines, as often as pos-
sible, the fiery cells that show in CT in the presence of yellow staining.
Nevertheless, MRI sweeps of the cerebrum of a patient contaminated with COVID-
19 give a lot of top to bottom detail in the delicate tissue than CT (Figure 11.12).
11.8.2 Advanced MRI for COVID-19 treatment

Using electromagnetic field and radiofrequency energy, an MRI test is adopted to
visualize the hydrogen atom density inside the human body. Generally, MRIs are
adopted to determine an issue on human joints, heart, cerebrum, wrists, ovary,

breast, bosoms, lower legs, and veins [51–56]. The radio wave communicated to a
collector in the machine converts the hydrogen population in our ROI to a 3D
picture. X-ray for imaging bloodstream does not include contrast-improving spe-
cialists; in any case, for the more exact imaging of organs that need a paramagnetic
metal complex, for the most part, gadolinium-based, which might be given orally or
intra-articularly.
These MRI contrast specialists planned remotely to tie especially in pathologic
tissue empowering improved infection discovery through changes in the R1 and R2
attractive unwinding rate constants of the neighborhood tissue.
The mix of at least two different upgrading specialists in a solitary MRI output
could give top-to-bottom analytic and prognostic data. Two different MRI contrast-
enhancing and high relaxive paramagnetic metal complexes are introduced for the
analysis. A multimodal CA with macromolecular blood-pool contrast-enhancing
properties as well as extravascular contrast-enhancing ability will provide more
information about the tumor and viral-infected cells [57].
In this case, a “theranostic” MRI contrast-enhancement complex is used to
explore the arrival of a beneficial molecule to the infection site, while the other is
used to audit restorative viability by providing a joint simultaneous voxel canny
evaluation of remedial delivery and reaction. On the other hand, two atomic MRI
contrast specialists could be joined to evaluate both columnist quality articulation
and the downstream impacts of the quality’s capacity, for example, synapse dis-
charge, protein production, or enzymatic activity. Similarly, a “dual-color image
enhancing” MRI strategy is adopted routinely to show a multi-viral specialist
imaging through fundamental fluorescence imaging experiments (Figure 11.13).
Additionally, this system will give more point-by-point data with a high spatial
goal, limitless imaging profundity, and 3D imaging abilities fundamental for non-
intrusive imaging of viral contamination, particularly in delicate tissues. Among
the enormous antiviral medications, coming up next are the examined ones for next
levels investigation. They chose to repurpose antiviral drugs chloroquine, hydro-
xychloroquine, ritonavir, lopinavir, and favipiravir. Twist hyperpolarization may
open new open doors in the biomedical conclusion of COVID-19 protein connec-
tion through MRI filtering methodology. Moreover, even if these medications up-
and-comers, which are under clinical examination, may not be helpful for the
COVID-19 conclusion [58]. This SABER-based hyperpolarization analysis has
been performed with high molecular weight protein and macro enzymes for early
diagnosing of “S” spike protein contamination in human organs and is given in
Figures 11.14 and 11.15.
The high molecular dimension CA is chosen from the viral database where the
drug has remarkable antiviral activity with fewer side effects. The antiviral drug
must be in current clinical application. This antiviral specialist drug was success-
fully analyzed with SABER-based hyperpolarization. The polarization over sig-
nificant distances was identified and tracked, implying that the advanced imaging
methodology can be adopted for other clinical analytic issues. By enhancing each
of the components adopted in this technique can be adapted to screen for the
Pre-contrast 10 min 20 min T1 (ms)

3,000
1,500
0
T2 (ms)
50
25
0
Pre-contrast 10 min 20 min (Gd) (mM)
0.15
0.075
0
(Dy) (mM)
0.7
0.35
Figure 11.13 Dual contrast agent [Gd(III) and Dy(III)] administered tumor
image from fluorescence MRI analysis
10
9 7
6 8
H N 10
N
2 8 7 7
4 5
1
Cl N
3
Hyperpolarization
12 34 5 6 7 8 9 10
9.5 9.0 8.5 8.0 7.5 7.0 6.5 6.0 5.5 5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0
1H NMR (ppm)
Figure 11.14 SABRE-treated hyperpolarized 1H NMR image of chloroquine drug

10 8 11
6 9 7
H N
N OH
2 9 8 8
4 5
1
Cl N Hyperpolarization
3
1 2 3 4 5 67 8 9 10, 11
10.0 9.5 9.0 8.5 8.0 7.5 7.0 6.5 6.0 5.5 5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5
1H NMR (ppm)
Figure 11.15 SABRE-treated hyperpolarized 1H NMR image of

hydroxychloroquine drug
dispersion and action of the repurposed antiviral medication, hereditarily planned

medication, or any new medication in vivo by MRI. It very well might be tackled to
additionally explore the subatomic correspondence of extra medication applicants
with mooring proteins and uncover unidentified activity on COVID-19, including
pharmacokinetics conduct.
11.9 Summary
Scientists, who won Nobel Prize for genetics, virology, drug designing, many other
live researchers, and doctors have confessed that the SARS-CoV-2 is a man-made
genetically modified virus. Scientists create numerous vaccines and antiviral drugs to
prevent the progression of the whole virus, but still it cannot be stopped. It is due to
the critical mechanism used in the construction of a viral molecule. The geneticists
who designed COVID-19 used three types of RNA, the majority of which came from
bats and pigs. Since RNA is important for constructing DNA in living things,
aggressive treatment on the virus-affected patient will lead to death. Therefore, the
only procedure we can do is to hide the RNA from coronavirus infection. Several
vaccines have been reported in the literature, but these are all accepted for medical
practices under pandemic situations without doing human trials. Therefore, the future
ill effect or the side effect of the currently available vaccine is unknown. Vaccines
will activate the human immune system against COVID-19 disease. However,
according to the WHO analysis and another medical report, there is widespread
concern that patients are dying as a result of SARS-CoV-2 virus infection even after
vaccination. On the other hand, it is recognized that the changeover in the protein
sequences from normal to high critical version on the coronavirus happens only
because of the vaccinated patient who got a COVID-19 infection.
Based on the discussion, it is concluded that each version of COVID-19
requires ACE2 enzyme for multiplication in the respiratory system, GI, kidney, and
brain, making it an important enzyme to recombine SARS-CoV-2 over RdRc. In
the spreading stage, the high polar region on the virus combined through RdRc, this
repeated procedure supports the nonstop viral RN acid replication. Furthermore, the
coarticulation of angiotensin-converting and TMPRSS2 proteins within the
respiratory system and stomach-related pathways have just been attributed as a key
for SARS-CoV-2 to enter the human tissue anatomy. The virus cannot generate
energy for its replication like cancer microbes. COVID-19 requires a huge amount
of sugars for their survival. This thirst for energy is satisfied by the presence of key
amino acids like Hi-296, Ar-345, Se-441, Ar-435, Se-436, Gly-464, and Ar-710 at
ACE2, and virus spike proteins. Therefore, if we need to stop the replication, we
need a target specific or amino acid-binding or anchoring antiviral drug. These
drugs can be newly designed or an already existing one. The requirement is the
high water solubility, long stay on the target, less steric strain, and zero side effects.
Currently, no antiviral drug has the abovesaid criteria. It is hoped that using in
silico methods, we will be able to design a new drug. Researchers can use in silico
methods to help them design new efficient drugs, find the anchoring spot over
coronaviruses, and test their antimicrobial properties. Researchers can use com-
puter programs such as Biorod, ChemSketch, and ACD labs to design drugs.
For evaluation, the molecular docking and molecular dynamics procedures are
addressed to be more user-friendly than the other algorithms. This computerized
algorithm will provide detailed amino acid sequences in viruses, enzymes, and any
complicated proteins, as well as additional information about the drug’s binding
efficiency and binding site on the coronavirus and other enzymes and proteins
involved in virus replication. For virtual screening, the AutoDock Vina and struc-
tural mechanics are found to be more reliable. The catalytic breaking interaction of
3CL-pro at distinct subsists of the polyproteins releases different polypeptides. As
discussed earlier, nearly 29 highly active proteins are involved in SARS-CoV-2
transmission and replication in the human lungs cell. 3CL-pro has been identified
as a potential-targeted therapy for COVID-19 diagnosis due to its vital role in
disease progression and indeed the lack of a near-human counterpart. According to
this molecular analysis of the M-pro, the Arg652 moiety is in the proper position to
form a link with Arg716 and Arg710, which are present over the ACE2 enzyme in
the human lung, brain, kidney, and GI tract. Any drug molecule that can disrupt the
linkage of these three amino acids will be the most effective anti-COVID drug. In
addition, it has been established that the M-pro protein is responsible for viral
replication as well as gene conversion or transmission. As a result, any drug
molecule that ideally ties the M-pro pocket may inhibit SARS-CoV-2 replication.
It is claimed that designing and developing a new chemical molecule with
these binding properties takes more time. Furthermore, their medicinal trial studies
on living organisms will take longer time. As a result, reanalyzing, reusing, or
restudying antibacterial and antiviral drugs against COVID infection will be a

better option in the current pandemic situation. These, too, require in silico methods
for evaluation. For this, many researchers select HIV, cancer, malarial, asthma, and
fungal drugs from FDA-approved viral drug data bank and antiviral databases.
Antiviral drugs, such as remdesivir, chloroquine, hydroxychloroquine, nitro-
derivatives of chloroquine, mitoxantrone, leucovorin, birinapant, and dynasore,
have been investigated utilizing docking and molecular modeling to identify
potential inhibitors of SARS-CoV-2 M-pro. According to the PCA results, the
unique s-profile-based attribute set “I” connects the primary COVID-19 drugs
remdesivir. In any case, the EIDD-2801 in monophosphate structures shows a more
vulnerable relationship with drugs that accounts for giving indications against
SARS-CoV-2 transmission. Alternatively, dual active drugs for COVID 19 and
other deadliest cancer drugs are also under animal trial study. This dual-model drug
is used in the most sophisticated CT and MRI scanning procedures. This advanced
research will provide information about infection sites, infection area or depth, and
live color image monitoring of drug movement inside the human body.
Our discussion highlights the applicability of machine-based strategies to find the

target site on SARS-CoV-2 that involves viral replication and to select an appropriate
drug to anchor the reactive site followed by the inhibition of viral replication with the
most accuracy and low cost. The current pandemic has not given much time to
undergo human trials I, II, and III, so we must rely on the existing drug whose effi-
ciency in both ends is known. The problem is still in the selection of antigens to inhibit
viral replication. To set right the pandemic in a fast and furious manner, we hope on in
silico methods. Every day, the researchers are giving additional instructions regarding
the viral protein structure, ACE2 cleavage pattern in lungs, GI tract, kidneys, and
brain, DNA contamination in human cells after COVID-19 infection, and the updated
activity of drugs under clinical trials. According to the study, in silico methods played
a significant role in detecting anchor sites on the SARS-main CoV-2’s protease and
another key amino acid that supports viral replication. The binding affinity and the
drug stability data from molecular docking and molecular dynamics study gives more
description about the drug synthesized so far and the various possible reactive sites for
drug targeting. Also, it is confirmed that a better potent drug must have the hydro-
phobic bulky group that can bind on the “anchor” site. The in silico predictions trigger
the researchers to discover multimodal inhibitors that can track the infection and kill
the target. The discussions also serve as a wake-up call to patients and clinicians to
conduct routine COVID-19 infection screenings even after the disease has been cured.
Finding the correct and efficient drug is the only way to eradicate this disease.
11.10.1 Future scope

Since there are more parallel computing techniques developed, the selection of an
appropriate procedure will be based on its suitability for analysis, result accuracy,
and time required. The ever-green PCR test itself makes the wrong decision on
COVID infection analysis. The PCR tool can be upgraded with a live gene repli-
cating algorithm to detect coronavirus infections and their structurally changed
virus molecule. Once the infection is confirmed, the next step is to locate the virus.
This information is critical for patients who have other chronic diseases as well as a
coronavirus infection. CT and MRI are the most promising treatments. The
advancement in the design of image-enhancing CAs with dual color-producing
properties will identify the locations where the virus spreads and the concentration
of the virus in the location. From this, the physician can understand the severity of
the organ, which got an infection. The biomolecular elemental analysis will give
information about the amino acid sequences, enzyme present, and tissue pH. Basic
information about the molecular sequence on the ROI is available in the Protein
Data bank. Next, it is in the researcher’s hand to analyze the chemical bonding
sequence through molecular imaging (AutoDock Vina-4.2) and analyzing program
(Gaussian algorithm integrated python module). For instance, the researcher can
select antiviral drugs like chloroquine, remdesivir, leucovorin, nelfinavir, and cur-
cumin from viral data bank and perform the docking and molecular dynamics study
to find the antiviral efficiency against SARS-CoV-2. Other synthetic scientists can
do their research in designing and evaluating drug molecules by using the pre-
viously discussed in silico procedures. Once the design is finished through this
algorithm, the binding energy and the viral inhibition efficiency can be calculated.
This will reduce the laboratory time and the cost of invention. The in vivo study can
be performed on animals and human samples using advanced MRI instruments.
The live monitoring of the drug action using CT scans and dual-color-based mul-
timodal MRI scanning agents will be the future scope of this study.
References
[1] Rome B. N. and Avorn J. ‘Drug evaluation during the Covid-19 pandemic’.
New English Journal of Medicine. 2020; 382(24): 2282–2284.
[2] Bang S., Son S., Kim S. et al. ‘Disease pathway cut for multi-target drugs’.
BMC Bioinformatics. 2019; 20(74): 1–12.
[3] Zhou J., Jiang X., He S., et al. ‘Rational design of multitarget-directed
ligands: Strategies and emerging paradigms’. Journal of Medicinal
Chemistry. 2019; 62(20): 8881–8914.
[4] Zhang L., Lin D., Sun X., et al. ‘Crystal structure of SARS-CoV-2 main
protease provides a basis for design of improved a-ketoamide inhibitors’.
Science. 2020; 368(6489): 409–412.
[5] Park M., Thwaites R. S., and Openshaw P. J. M. ‘COVID-19: Lessons from
SARS and MERS’. European Journal of Immunology. 2020; 50(3): 308–311.
[6] Rabaan A. A., Al-Ahmed S. H., Haque S., et al. ‘SARS-CoV-2, SARS-CoV,
and MERS-COV: A comparative overview’. Le Infezioni in Medicina. 2020;
28(2): 174–184.
[7] Shanmugaraj B., Siriwattananon K., Wangkanont K., and Phoolcharoen W.
‘Perspectives on monoclonal antibody therapy as potential therapeutic
intervention for Coronavirus disease-19 (COVID-19)’. Asian Pacific

Journal of Allergy and Immunology. 2020; 38(1): 10–18.
[8] Gao Y., Yan L., Huang Y., et al. ‘Structure of the RNA-dependent RNA
polymerase from COVID-19 virus’. Science. 2020; 368(6492): 779–782.
[9] Tai W., He L., Zhang X., et al. ‘Characterization of the receptor-binding
domain (RBD) of 2019 novel coronavirus: Implication for development of
RBD protein as a viral attachment inhibitor and vaccine’. Cellular &
Molecular Immunology. 2020; 17: 613–620.
[10] Basu A., Sarkar A., and Maulik U. ‘Molecular docking study of potential
phytochemicals and their effects on the complex of SARS-CoV2 spike
protein and human ACE2’. Scientific Reports. 2020; 10: 17699.
[11] Wu C. I. and Wen H. ‘Heightened protein-translation activities in mamma-
lian cells and the disease/treatment implications’. National Science Review.
2020; 7: 1847–1851.
[12] Hoffmann M., Kleine-Weber H., Schroeder S., et al. ‘SARS-CoV-2 cell
entry depends on ACE2 and TMPRSS2 and is blocked by a clinically proven
protease inhibitor’. Cell. 2020; 181: 271–280.
[13] Lanjanian H., Moazzam-Jazi M., Hedayati M., et al. ‘SARS-CoV-
2 infection susceptibility influenced by ACE2 genetic polymorphisms:
Insights from Tehran Cardio-Metabolic Genetic Study’. Scientific
Reports. 2021; 11: 1529.
[14] Can H., Koseoglu A. E., Erkunt Alak S., et al. ‘In silico discovery of anti-
genic proteins and epitopes of SARS-CoV-2 for the development of a vac-
cine or a diagnostic approach for COVID-19’. Scientific Reports. 2020; 10:
22387.
[15] Shaddel M., Ebrahimi M., and Tabandeh M. R. ‘Bioinformatics analysis of
single and multi-hybrid epitopes of GRA-1, GRA-4, GRA-6 and GRA-7
proteins to improve DNA vaccine design against Toxoplasma gondii’.
Journal of Parasitic Diseases. 2018; 42(2): 269–276.
[16] Zheng M., Karki R., Williams E. P. et al. ‘TLR2 senses the SARS-CoV-2
envelope protein to produce inflammatory cytokines’. Nature Immunology.
2021; 22: 829–838.
[17] Hou H., Wang T., Zhang B., et al. ‘Detection of IgM and IgG antibodies in
patients with coronavirus disease 2019’. Clinical & Translational
Immunology. 2020; 9(5): e01136.
[18] Chakraborty C. ‘Performance analysis of compression techniques for
chronic wound image transmission under smartphone-enabled tele-wound
network’. International Journal of E-Health and Medical Communications
(IJEHMC). 2019; 10(2): 1–15, 10.4018/IJEHMC.2019040101.
[19] Chakraborty C. ‘Chronic wound image analysis by particle swarm optimi-
zation technique for tele-wound network’. International Journal of Wireless
Personal Communications. 2017; 96(3): 3655–3671, ISSN: 0929-6212,
10.1007/s11277-017-4281-5.
[20] Joffre O. P., Segura E., Savina A., and Amigorena S. ‘Cross-presentation by
dendritic cells’. Nature Reviews Immunology. 2012; 12(8): 557–569.
[21] Zhu X., Wang X., Han L., et al. ‘Multiplex reverse transcription loop-
mediated isothermal amplification combined with nanoparticle-based lateral
flow biosensor for the diagnosis of COVID-19’. Biosensors and
Bioelectronics. 2020; 166: 112437.
[22] Corman V. M., Landt O., Kaiser M., et al. ‘Detection of 2019 novel cor-
onavirus (2019-nCoV) by real-time RT-PCR’. Eurosurveillance. 2020; 25
(3): p2000045.
[23] Shen Z., Xiao Y., Kang L., et al. ‘Genomic diversity of severe acute
respiratory syndrome coronavirus 2 in patients with coronavirus disease’.
Clinical Infectious Diseases. 2020; 71(15): 713–720.
[24] Smyrlaki I., Ekman M., Lentini A., et al. ‘Massive and rapid COVID-19 testing
is feasible by extraction-free SARS-CoV-2 RT-PCR’. Nature Communications.
2020; 11: 4812.
[25] Abduljalil J. M. ‘Laboratory diagnosis of SARS-CoV-2: Available approa-
ches and limitations’. New Microbes and New Infections. 2020; 36: 100713.
[26] Kevadiya B. D., Machhi J., Herskovitz J., et al. ‘Diagnostics for SARS-CoV-
2 infections’. Nature Materials. 2021; 20: 593–605.
[27] Trott O. and Olson A. J. ‘AutoDock Vina: Improving the speed and accuracy
of docking with a new scoring function, efficient optimization and multi-
threading’. Journal of Computational Chemistry. 2010; 31(2): 455–461.
[28] Tien H., Haoran W., and Binquan L. ‘In silico exploration of the molecular
mechanism of clinically oriented drugs for possibly inhibiting SARS-CoV-
2’s main protease’. The Journal of Physical Chemistry Letters. 2020; 11(11):
4413–4420.
[29] Guedes I. A., Costa L. S. C., dos Santos K. B., et al. ‘Drug design and
repurposing with DockThor-VS web server focusing on SARS-CoV-2 ther-
apeutic targets and their non-synonym variants’. Scientific Reports. 2021; 11
(1): 5543.
[30] Keretsu S., Bhujbal S. P., and Cho S. J. ‘Molecular modeling studies of
pyrrolo [2, 3-d] pyrimidin-4-amine derivatives as JAK1 inhibitors based on
3D-QSAR, molecular docking, molecular dynamics (MD) and MM-PBSA
calculations’. Journal of Biomolecular Structure and Dynamics. 2021; 39(3):
753–765.
[31] Pushpakom S., Iorio F., Eyers P. A., et al. ‘Drug repurposing: Progress,
challenges and recommendations’. Nature Reviews Drug Discovery. 2019;
18(1): 41–58.
[32] Chen Y. W., Yiu C. P. B., and Wong K. Y. ‘Prediction of the SARS-CoV-2
(2019-nCoV) 3C-like protease (3CL pro) structure: Virtual screening reveals
velpatasvir, ledipasvir, and other drug repurposing candidates’.
F1000Research. 2020; 9: 129.
[33] Muralidharan N., Sakthivel R., Velmurugan D., and Gromiha M. M.
‘Computational studies of drug repurposing and synergism of lopinavir,
oseltamivir and ritonavir binding with SARS-CoV-2 protease against
COVID-19’. Journal of Biomolecular Structure and Dynamics. 2021; 39(7):
2673–2678.
[34] Muhammad L. J., Algehyne E. A., Usman S. S., Ahmad A., Chakraborty C.,
and Mohammed I. A. ‘Supervised machine learning models for prediction of
COVID-19 infection using epidemiology dataset’. SN Computer Science.
2021; 2(11): 1–13.
[35] Elfiky A. A. ‘Ribavirin remdesivir, sofosbuvir, galidesivir, and tenofovir
against SARS-CoV-2 RNA dependent RNA polymerase (RdRp): A mole-
cular docking study’. Life Sciences. 2020; 253: 117592.
[36] Wu C., Liu Y., Yang Y., et al. ‘Analysis of therapeutic targets for SARS-
CoV-2 and discovery of potential drugs by computational methods’. Acta
Pharmaceutica Sinica B. 2020; 10(5): 766–788.
[37] Assis L. C., de Castro A. A., de Jesus J. P. A., et al. ‘Computational evidence
for nitro derivatives of quinoline and quinoline N-oxide as low-cost alternative
for the treatment of SARS-CoV-2 infection’. Scientific Reports. 2021; 11: 6397.
[38] Murugan N. A., Kumar S., Jeyakanthan J., and Srivastava V. ‘Searching for
target-specific and multi-targeting organics for Covid-19 in the Drug bank
database with a double scoring approach’. Scientific Reports. 2020; 10: 19125.
[39] Lu H. ‘Drug treatment options for the 2019-new coronavirus (2019-nCoV)’.
BioScience Trends. 2020; 14(1): 69–71.
[40] Liu X., Li Z., Liu S., et al. ‘Therapeutic effects of dipyridamole on COVID-
19 patients with coagulation dysfunction’. Acta Pharmaceutica Sinica B.
2020; 10(7): 1205–1215.
[41] Van Zundert G. C. P., Rodrigues J. P. G. L. M., Trellet M., et al. ‘The
HADDOCK 2.2 webserver: User-friendly integrative modeling of biomole-
cular complexes’. Journal of Molecular Biology. 2016; 428(4): 720–725.
[42] Kumar N. ‘Cytotoxic T-lymphocyte elicited vaccine against SARS-CoV-2
employing immunoinformatics framework’. Scientific Reports. 2021; 11: 7653.
[43] Elmezayen A. D., Al-Obaidi A., Şahin A. T., and Yelekçi K. ‘Drug repur-
posing for coronavirus (COVID-19): In silico screening of known drugs
against coronavirus 3CL hydrolase and protease enzymes’. Journal of
Biomolecular Structure and Dynamics. 2020: 1–13.
[44] Garg L., Chukwu E., Nasser N., Chakraborty C., and Garg G. ‘Anonymity
model’. IEEE Access. 2020; 8: 159402–159414.
[45] Chakraborty C., and Abougreen A. N. ‘Intelligent Internet of Things and
Transactions on Pervasive Health and Technology. 2021: 1–14.
[46] Taka E., Yilmaz S. Z., Golcuk M., et al. ‘Critical interactions between the
SARS-CoV-2 spike glycoprotein and the human ACE2 receptor’. The
Journal of Physical Chemistry B. 2021; 125(21): 5537–5548.
[47] Choy K. T., Wong A. Y., Kaewpreedee P., et al. ‘Remdesivir, lopinavir,
emetine, and homoharringtonine inhibit SARS-CoV-2 replication in vitro’.
Antiviral Research. 2020; 178: 104786.
[48] Bullen C. K., Hogberg H. T., Bahadirli-Talbott A., et al. ‘Infectability of
human BrainSphere neurons suggests neurotropism of SARS-CoV-2’. Altex.
2020; 37(4): 665–671.
[49] Bernheim A., Mei X., Huang M., et al. ‘Chest CT findings in coronavirus
disease-19 (COVID-19): Relationship to duration of infection’. Radiology.
2020; 295(3): 200463.
[50] Zhou Z., Guo D., Li C., et al. ‘Coronavirus disease 2019: Initial chest CT
findings’. European Radiology. 2020; 30(8): 4398–4406.
[51] Pushparaj L. T. and Alexander V. ‘Trinuclear Gd(III) metal complex with
amide core display remarkable enhancement in relaxivity’. Applied
Magnetic Resonance. 2017; 48(8): 813–825.
[52] Salarian M., Turaga R. C., Xue S., et al. ‘Early detection and staging of
chronic liver diseases with a protein MRI contrast agent’. Nature
Communications. 2019; 10: 4777.
[53] Pushparaj L. T., Francy I. R. E., and Uma D. M. ‘HSA and CA-125 binding
study of [Pr-(DO3-Ch-Ph-Am-Gd(III))2Pt(IV)] complex as M-MRI contrast
agent for ovarian cancer treatment’. Journal for Innovative Development in
Pharmaceutical and Technical Science. 2021; 4: 1–8.
[54] Anderson C. E., Donnola S. B., Jiang Y., et al. ‘Dual contrast – Magnetic
resonance fingerprinting (DC-MRF): A platform for simultaneous quantifi-
cation of multiple MRI contrast agents’. Scientific Reports. 2017; 7: 8431.
[55] Lindeman L. R., Jones K. M., High R. A., et al. ‘Differentiating lung cancer
and infection based on measurements of extracellular pH with acidoCEST
MRI’. Scientific Reports. 2019; 9: 13002.
[56] Abbasi A., Javed A. R., Chakraborty C., Nebhen J., Zehra W., and Jalil Z.
‘ElStream: An ensemble learning approach for concept drift detection in
dynamic social big data stream learning’. IEEE Access. 2021: 1–12.
[57] Kevadiya D., Machhi J., Herskovitz J., et al. ‘Diagnostics for SARS-CoV-
2 infections Bhavesh’. Nature Materials. 2021; 20:593–605.
[58] Anderson C. E., Johansen M., Erokwu B. O., et al. ‘Dynamic, simultaneous
concentration mapping of multiple MRI contrast agents with dual contrast –
Magnetic resonance fingerprinting’. Scientific Reports. 2019; 9(1): 19888.
Chapter 12
Stress detection for cognitive rehabilitation in
COVID-19 scenario
Ahona Ghosh1, Sima Das1 and Sriparna Saha1
Abstract
Due to the current demand for emerging technologies like the Internet of Things
integrated with machine learning in industry and academics, brain–computer
interface tools like electroencephalogram in healthcare have drawn worldwide
attention. As has been noticed that during recent times, mobile phone exposure to
people increased in at least 2-fold way, so games have been used as stimuli for
detecting how our brain becomes overburdened with increased exposure. After the
data acquisition from 14 channels of an electroencephalogram, the activated
regions were identified. Features were extracted from the most activated ten elec-
trode channels using discrete wavelet transform. To reduce the dimensions of the
feature space for enhancing the performance, principal component analysis was
used. The mental state classification was performed using a support vector machine
based on the detected stress. The proposed system has outperformed the existing
ones for its effectiveness and efficiency in a broad application area of cognitive
rehabilitation. Classification accuracy was obtained as 92.79% and different other
metrics proved that the combination of channel selection, feature extraction, and
classification methods in our proposed approach has outperformed the others.
Privacy is maintained, and it is flexible to the user as per his/her convenient time.
Keywords: Electroencephalogram; Stress detection; COVID-19 pandemic;

Discrete wavelet transform; Dimensionality reduction; Support vector
machine; Cognitive rehabilitation
12.1 Introduction
In this era, stress is a common event that happens to everyone. Machine learning
(ML) is set as a benchmark in emerging technology, and EEG is connected with the
1
Department of Computer Science and Engineering, Maulana Abul Kalam Azad University of
Technology, West Bengal, India
brain–computer interface for collecting electrical signals and converts it into a

numerical dataset to measure human brain activity. The recent pandemic due to
COVID-19 had a tremendous impact on human psychological health [1,2]. In our
daily life, many of us are facing challenges during the COVID-19 pandemic era.
The challenges are as follows: physical health issues and mental stress. The
undertaken research on previous outbreaks caused by Middle Eastern respiratory
syndrome [3], severe acute respiratory syndrome [4], and Ebola [5] showed dif-
ferent consequences of a burden on mental health, like depression, panic, anxiety,
and stress pandemics [6,7].
Mental stress increases by overwhelming, feeling of fear, chronic health pro-
blems, feeling lonely, and isolated from family members. It is natural to feel
stressed in a pandemic era, but it is injurious for mental as well as physical health
also. The recent pandemic due to COVID-19 has impacted the quality of our life-
style and well-being with multiple risk factors [8]. While most of the recent studies
have focused on the psychological effects of pandemics only on students and aca-
demic staff, this chapter has attempted to detect the stress among ordinary people
by detecting abnormality or degradation in the brain activities recorded by
electroencephalogram.
Artificial intelligence and intelligent systems in smart healthcare especially
during the pandemic are experiencing an increasing demand [9] to read and act on
complex data, such as photos, speech, text, video, or other signals, to achieve a
required goal. Blockchain also has been widely used to trace private contacts of
Internet of Medical Things framework users digitally, instead of physical contact in
the COVID-19 scenario [10]. Using computed tomography datasets and chest X-ray
images, CNN-based pretrained models have been widely employed for COVID-19
detection. However, for training, validation, and testing, most of the approaches
utilized a smaller number of data samples for both CT and CXR datasets. As a result,
while the model may have performed well during testing, it will not be more effec-
tive in the long run [11]. The importance of diagnostics as a cornerstone for making
life-saving decisions was demonstrated in this dramatic case [12,13].
Our proposed work can be used to detect human stress in a pandemic. We have
discussed how psychological interventions affect the human mind during the
increased game-playing habit of the pandemic. We have used electro-
encephalography to collect datasets from the human brain and generate a brain map
to detect which lobe is more active. After identifying the lobe positions, AF3, AF4,
T8, T7, P8, P7, O2, O1, FC6, and FC5 channels are selected from the brain map
generated by sLORETA software. Based on this, feature extraction is carried out
using discrete wavelet transform (DWT). After that, classification is performed
using a support vector machine (SVM) for the recognition of the subject’s mental
state as stressed or not stressed. Performance evaluation metrics, namely recall,
error rate, precision, F1 score, and accuracy, have been used to assess the efficiency
and applicability of our proposed approach to detect stress among game players in
real-time scenarios.
There have been several attempts toward cognitive rehabilitation using ML
techniques. The gaps found in state of the art have motivated us to consider this
Stress detection for cognitive rehabilitation 333
area for further exploration. The motivations behind taking up this research chal-
lenge include the following:
1. to study the presence of stress in players based on android games;
2. to overcome the shortcoming present in state-of-the-art literature and to gra-
dually improve the performance of stress detection;
3. to detect the three factors of game-based regulating mental state, i.e., related-
ness (the capability of gamers to other people in society), the competence of
completing necessary game levels, and autonomy.
The contributions of our proposed framework for the detection of stress levels
in ordinary people during the pandemic era can be briefed as per the following:
1. It has been able to detect whether the subject is stressed or not stressed.
2. Haar as the wavelet basis function in the DWT used as the feature extractor in
our proposed approach has outperformed the other wavelet basis functions
used in the state-of-the-art literature.
3. The training set size is larger than the existing benchmarks in the related area
as per our knowledge. So, the classification accuracy of the SVM model has
been achieved as 92.79%, which outperforms the state-of-the-art literature in
its concerned domain. Also, hyperparameter tuning has been carried out to
choose the best parameters for achieving the highest accuracy.
The remaining of the chapter is as follows: in Section 12.2, we have discussed
related works of stress level detection and cognitive load; in Section 12.3, we have
proposed our system and algorithm of the system also be discussed; experimental
outcomes are discussed in Section 12.4; last but not the least, concluding statements
regarding the proposed approach are given in Section 12.5.
12.2 Related works

Rehabilitation is the art of restoring someone from damages or disorders from
mental and physical aspects. Two therapeutic seal robots have been used in [14] to
achieve rehabilitation for dementia patients by recording patient’s EEG data 20 min
before and after the therapy. Patients who liked the seal robots and interacted with
them freely were observed to improve cortical neurons activity. However, the
durability and repeatability of the positive effects are to be investigated further to
find the efficacy of the model in a real-world scenario. Excited, happy, pleased,
relaxed, calm, distressed, frightened, angry, depressed, and sad are the emotions
considered by Ali et al. while recognizing sentiments by EEG [15]. Statistical
features, wavelet entropy, modified energy, and wavelet energy features are the
determining factor of these emotions, and the average accuracy obtained from three
classifiers, namely SVM, k-nearest neighbor (kNN), and quadratic discriminant
analysis, was 83.87%. A minimum number of electrodes, i.e., only four were used
to collect data from the EEG device that proves the vital role of the prefrontal
cortex in regulating and identifying emotions based on the requirement.
Alpha, beta, gamma1, gamma2, delta, theta band power, corresponding

asymmetry, and paired symmetry are the linear features, and detrended fluctuation
analysis, sample entropy are the nonlinear features considered by Mahato and Paul
in depression detection and severity-scaling application of a six-channel EEG [16].
After selecting features by ReliefF, bagging of SVM was used along with three
kernel functions, namely sigmoidal, Gaussian, and polynomial to classify the
depression intensity among which Gaussian kernel achieved the highest accuracy
and temporal region was proved to be responsible and affected for depressing
emotions generated in our brain. Data acquisition from only six channels of the
electrode made the framework easier to use and more convenient for common
people diagnosing depression. Linear discriminant analysis, genetic algorithm, and
decision tree (DT) have been applied to extract and select the most relevant EEG
features and differentiate between depressive and non-depressive subjects in [17].
Classification accuracy and optimality remarkably increased during the analysis of
the combined and complete set of features rather than separate analysis of indivi-
dual bands. EEG-based antidepressant response index has been used to detect the
treatment response from depressed subjects.
Kernel eigen-filter-bank common spatial pattern has been applied in feature
extraction and frequencies ranging between gamma and theta bands have been
selected in [18], after which, principal component analysis (PCA) has reduced the
feature vector dimension. Single-trial classification accuracy was achieved as 81%.
The participant-independent accuracy of the nonlinear SVM classifier was also
tested by a condition of leaving one participant based on voting in every iteration
throughout the classification process. The experimental setup consisted of only
eight electrodes, designing which is less time-consuming and the experiment lasted
only for 6 s which also indicates no exhaustion experience during the test data
collection. However, the system got tested with only 24 participants, which needs
an increase in training size to validate it further in real-time scenarios. Moreover,
the other regions apart from the temporal lobe can be tested further during channel
selection with methods like mutual information, Fisher’s separability criterion.
After providing a questionnaire on their previous experience of depression and
treatment with antidepressant medication to 21 subjects in [19], Niemiec and
Lithgow have identified that people having depression experience in past tend to
get a frontal brain asymmetry ratio lying near the distribution boundaries, and the
clear alpha peak helped to make the assessment procedure more consistent and
robust. After the noise removal using finite impulse response method along with
adaptive predictor filter, discrete wavelet transformation, and Kalman derivation,
the minimal redundancy maximal relevance method got applied to select features in
Cai et al.’s depression identification approach using a three-electrode pervasive
EEG [20]. Ten-fold cross-validation was applied to test the performance of SVM,
artificial neural network, kNN, and classification tree that resulted in the highest
accuracy of kNN as 79%. The theta wave absolute power worked most efficiently
in the learning process of discriminating depressed subjects from the normal ones.
Interhemispheric asymmetry and band power as linear features alone, wavelet
entropy and relative wavelet energy as nonlinear features alone, and a combination
of both got considered to classify healthy and depressed subjects using quadratic
and linear discriminant analysis, multilayer perceptron, and radial basis function
(RBF) network in [21]. The alpha band as the linear feature and both of the non-
linear features achieved the highest accuracy with a multilayer perceptron, RBF,
and linear discriminant analysis, respectively. The combined linear and nonlinear
feature set performed best with the RBF in terms of specificity, sensitivity, and
accuracy. However, a more generalized outcome can be achieved by testing the
framework with a greater number of patients in the future. Among the five layers of
CNN, the first four max-pooling layers were used in feature selection and the last
one, i.e., fully connected layer was applied in a classification where batch nor-
malization and dropout also took place to reduce the chance of overfitting in [22].
Here EEG data obtained from only two electrodes achieved satisfactory accuracy
since the training sample size of the PhysioNet database was also effective for such
a motor imagery task and CNN alone has extracted and selected features; thus, the
complexity of employing other artificial methods to preprocess, and feature
extraction was avoided also.
Artificial intelligence has been widely applied in different COVID-19-based
research works [23]. The existing literature on COVID-19 somewhere has
attempted to model the lockdown decision-making [24] and somewhere has
attempted to detect respiratory distress due to virus attack [25]. Additionally, the
convolutional network has verified its efficiency in various fields, including sen-
timent analysis [26], gesture recognition [27], sign language detection [28].
Table 12.1 presents a comparative study among state-of-the-art works from where
it is clear that existing literature lacks suitable infrastructure for stress detection and
recovery systems and as per our understanding, the proposed framework has
addressed the gaps found in those. The impact of playing different games and
beating the other player’s scores on human emotion has been analyzed in [29] using
EEG. A deep learning-based type 2 fuzzy set has been applied to classify the
emotion changes during game playing. Facial expression as well as brain signals
were monitored parallelly, and the games for which the negative emotions domi-
nated the positive ones can be avoided for a better quality of life in the future [30].
A bidirectional long short-term memory (LSTM) has been used based on the
player’s attention while engaged in game playing, and the dimensionality-based
scarcity has been handled using feature augmentation technique before feeding it in
the LSTM for emotion classification in [31]. The system was validated with some
combined five-by-two paired t-test and Friedman test. Based on the rating of
adversarial effects, the spreading of damaging computer and android games may be
controlled by the parents.
12.3 Proposed framework

Here we discuss the different aspects of our proposed methodology. After the
description of data acquisition techniques, the working mechanisms of feature
extraction and selection [42] are described at first, then the mental state
Table 12.1 A comparative study among state-of-the-art literature
Ref. de- Motivation EEG device Subject Preprocessing Feature extrac- Classification/ Contribution Shortcoming
tail tion Clustering found (if any)
Asif Stress level Four-channel A total of Notch band stop Absolute power, Sequential mini- The highest accu- The training sam-
et al. detection in MUSE 13 fe- filter rejectingrelative mal optimiza- racy (98.76% ple can be in-
[32] response to head-band males frequency be- power, coher- tion, stochastic for two-level creased in the
the music and tween 45 and ence, phase gradient des- and 95.06% for number of sub-
track 14 males 64 Hz lag, and am- cent, logistic three-level) jects and fea-
plitude asym- regression, was obtained tures in the
metry multilayer per- from logistic future to get
ceptron regression higher accuracy
using deep
learning
Saeed Long-term EMOTIV Xa- A total of DC offset re- Power spectral Support vector Involvement of The training sam-
et al. stress re- vier test- 13 fe- moval using density of five machine, naı̈ve psychology ex- ple can be in-
[33] cognition in bench ver- males fast Fourier channels, al- Bayes, k near- perts to label creased in
resting-state sion and transform and pha and beta est neighbor, the stress in- several subjects
by PSS-10 3.1.21 head- 20 males minimal eye- asymmetry, multilayer per- tensity and and features in
question- set blink in closed and feature ceptron, logis- highest accu- the future to get
naire eye state selection tic regression racy achieved higher accuracy
using t-test by SVM and using deep
LR having al- learning mod-
pha symmetry els
feature
Ahn Stress assess- Two-channel 14 male EOG signal re- Right and left Analysis of var- The combination The convenience
et al. ment using EEG and subjects moval using normalized al- iance to con- of EEG and to put on such a
[34] mental ar- one channel only fast Fourier pha and beta firm the heart rate device can be
ithmetic, ECG transform band power, statistical sig- variability im- increased by
Stroop col- and each nificance and proved the replacing the
or–word band’s power support vector stress assessing wet electrode
test asymmetry machine to classifier per- with dry ones
classify stress formance and the training
situations sample size can
be increased for
better accuracy
Kalas Stress level RMS MAXI- Ten sub- Noise reduction Features from Calculation of Helpful in stress Sensitivity analy-
and estimation MUS jects only using discrete two stages, stress indices management sis and inter-
Mo- and reduc- 24 channel wavelet trans- i.e., baseline and grouping and takes less vention can be
min tion EEG form and task-load into high stress time and hu- considered a
[35] got extracted and low stress man resource future exten-
based on a sion to check
threshold using stress reduction
K means clus-
tering
Priya Stress detec- EEG having a A total of Artifact-free EEG A total of 64 SVM and k-near- Gaussian SVM More evaluation
et al. tion during 10–20 elec- 36 sub- signal was power ratio est neighbor as having KS>2.5 on higher order
[36] serial sub- trode loca- jects considered feature extrac- a classifier and performed bet- polynomial
traction tion system containing 182 tion by power k-fold cross- ter. kNN with SVM can be
task s of relaxed spectral den- validation to one neighbor done in the fu-
state and 62 s sity using fast check the gen- performed well ture as it could
of a stressed Fourier trans- eralized and in that work achieve good
state form unbiased per- performance
formance but takes more
time
Liao Stress predic- NeuroSky Seven sub- Not mentioned EEG frequency Binary classifica- The highest vali- Click button ad-
et al. tion by lis- Mind-wave jects only using Fast tion into two dation accuracy dition can be
[37] tening to Mobile Fourier Trans- classes, i.e., is 80.13%. done in the fu-
music and form meditation and High precision ture to control
calm the attention using for meditation the action
mind down deep learning class, better where higher
model with F1-score, and attention means
ReLU activa- high recall for higher button
tion function attention class. click accuracy
(Continues)
Ref. de- Motivation EEG device Subject Preprocessing Feature extrac- Classification/ Contribution Shortcoming
tail tion Clustering found (if any)
and fully con- Small MSE
nected layers loss of 0.0882
Sharma Stress identi- Not men- Not men- Fifth-order But- EEG signal de- Support vector The highest accu- The training sam-
and fication tioned tioned terworth filter composition machine used racy in the al- ple size is not
Cho- during the has outper- into intrinsic in the proposed pha frequency mentioned and
pra aptitude test formed the mode function model has out- band is 94% integration of
[38] combined 45- and frequency performed k and the average deep learning
Hz finite im- feature extrac- nearest neigh- is 90%. The method can be
pulseResponse tion using bor, quadratic accuracy of incorporated in
(FIR) filter and Hilbert– discriminant SVM is the the future for
0.75 Hz high Huang trans- analysis, and highest greater perfor-
pass filter form linear discrimi- mance
nant analysis
Peng Chronic stress EEG of a 11 Artifact removal A total of 20 Statistical analy- Features had a The number of
et al. detection 10–20 elec- using low pass features sis using Ana- significant dif- control and ex-
[39] and moni- trode loca- filter of 40 Hz namely max lysis of ference for the perimental
toring tion system cut off fre- power, abso- variance in control and group is not the
quency and lute power, SPSS 17.0 and stressed groups same, thus,
EOG removal relative paired sample in nonlinear more samples
using wavelet power, center t-test to analyze EEG analysis. should be
method frequency, en- and differenti- The only rela- added. Sensi-
tropy, correla- ate between left tive power of tivity and spe-
tion dimen- and right hemi- alpha-band was cificity should
sion, alpha sphere activ- significantly be used to
asymmetry ities of the different for evaluate the
brain control and performance
stressed in
linear EEG
analysis
Virmani Stress recog- 10–20 elec- 23 Not mentioned Two emotional Russell’s circum- The overall accu- Accuracy is not
et al. nition using trode place- features, plex model- racy of k near- up to the mark
[40] ECG and ment sys- namely arou- based k nearest est neighbor— which can be
EEG for tem EEG sal and va- neighbor and 68% and deci- increased by
audio and lence using decision tree to sion tree—75% having more
visual sti- power spectral classify be- training sample
muli density tween stressed size and other
and non- models can be
stressed sub- tested for better
jects performance
Al- Stress quanti- BrainMaster A total of Third-order But- Wavelet trans- Two-sample t-test Accuracy of SVM The right prefron-
Shar- fication 24E system 12 sub- terworth filter form for five having mean p- decreased with tal cortex has
gie during three jects only in passband decomposition values as 0.03, increasing level been identified
et al. levels of frequency levels and one 0.042, and 0.05 of the given ar- as the responsi-
[41] arithmetic 0.5–30 Hz and approximation for level one, ithmetic task ble brain region
task artifact re- level ranging two, and three, for stress; how-
moval using between 0 and respectively ever, the result
independent 30 Hz can be im-
component proved further
analysis by considering
a larger training
size
classification using SVM’s workflow has been presented. The functionality of our
entire system has been pictorially presented in Figure 12.1.
12.3.1 Introduction to EEG

Electroencephalogram abbreviated as EEG has been used as the data acquisition
device in our proposed framework [43]. Different regions of the human brain
responsible for different activities and the electrode placement based on the inter-
national 10–20 system have been depicted in Figure 12.2. Here the 10 and 20 mean
the distance between adjacent nodes is 10% or 20% of the entire scalp. Every
electrode is represented by a letter followed by a digit where the letters F, P, O, C,
and T stand for the frontal, posterior, occipital, central, and temporal region,
respectively, and the digit is odd representing the left hemisphere and even repre-
senting the right one. N and I denote Nasion and Inion, respectively [44].
Data acquisition Feature extraction Classification using support Recognition of mental

using EEG using discrete wavelet vector machine state
transform
Feeling
Stressed Not stressed
Cognitive
rehabilitation
Thoughts Behavior
Figure 12.1 Illustration of our proposed framework applied in rehabilitation
AF3 AF4
F7 F8
F3 F4
FC5 FC6
T7 T8
P7 P8
O1 O2
Figure 12.2 EEG electrodes used to acquire data in the proposed approach
12.3.2 Feature extraction using DWT

The wavelet transform concept is the representation or approximation of functions
or signals using a function group, called a wavelet function system. Wavelets show
two elementary properties: scale and location. The first one, also called dilation,
defines the stretch or squish of a wavelet, which is related to the frequency of
waves. Location describes the position of the wavelet-based on time or space [45].
Among the two types of the wavelet transform techniques, the continuous wavelet
transform (CWT) considers all the probable wavelets over an unlimited number of
scales and locations, whereas the DWT makes use of a limited set of wavelets,
which are distinct at certain scales and locations. Here lies the main distinction
between these two versions. A wavelet can be represented by the first-order deri-
pffiffiffiffiffiffiffiffiffi
vative of Gaussian function ðs nÞeððsnÞ =2m Þ= 2Pm where the wavelet scale
2 2 3
and location are represented by a and b, respectively. The CWT of signal s(t) can be
defined by
ð þ1
Tc ðm; nÞ ¼ sðtÞgðm;nÞ ðtÞdt (12.1)
1
where the continuous wavelet basis function is

1 ðt nÞ
gðtÞm;n ¼ pffiffiffiffi g (12.2)
m m
for which the categories are Haar, db series, coiflet, and so on. Since the EEG
signals are discrete, thus DWT has been used in our proposed approach to extract
features from it. Compared with the CWT, the DWT limits the m and n of the
wavelet basis function g(m, n) to discrete points, which is called the scale dis-
cretization and displacement. The DWT of signal s(t) is defined by
1 X 1 X 1 X
Td ðtÞ ¼ pffiffiffiffi Wx ðp0 ; qÞxp0 ;q ðtÞ þ pffiffiffiffi Wy ðp; qÞyp;q ðtÞ (12.3)
m k m p¼p0 k
where the approximate or scaling coefficient is
1 X m1
Wx ðp0 ; qÞ ¼ pffiffiffiffi Td ðtÞ~x p0 ;q ðtÞ: (12.4)
m t¼0
The detailed coefficient is
1 X m1
Wy ðp; qÞ ¼ pffiffiffiffi Td ðtÞ~y p;q ðtÞ (12.5)
m t¼0
and p0 is the arbitrary initial scale. Figure 12.3 has shown the mechanism of
wavelet decomposition of a signal s(t) by considering three layers of decomposition
for an instance [46] where x and y have been used for the approximate and detail
coefficient, respectively.
12.3.3 Feature selection using principal component

analysis
Noise regarding eyeblinking artifacts has been removed in our proposed archi-
tecture using PCA that is a popular dimension reduction method in ML. The
working mechanism of PCA in noise removal has been shown in Figure 12.4. PCA
can be described as an orthogonal linear transformation that translates data into a
new coordinate system so that the largest variance by some scalar projection of the
data falls on the first coordinate (named PC1), the second major variance on the
second coordinate (named PC2), etc. [47].
If we consider a D-dimensional data matrix M of r and (cþ1) number of rows
and columns where every column among c number of columns agrees to an inde-
pendent variable of matrix X and the remaining one column corresponds to a
dependent variable denoted by Y, the step-by-step procedure to perform PCA in this
matrix to obtain reduced dimensionality d where d D is shown in the following
algorithm.
s(t)
x(t) y(t)
2 2
x(t) y(t) x (t) y (t)
2 2 2 2
x(t) y(t) x(t) y(t) x(t) y (t) x(t) y(t)
Figure 12.3 Three-layer wavelet decomposition in our proposed framework
x1
PC1
PCA PC1 (Maximum

variance)
PC2 PC2
(Second maximum variance)
0 x2
Original three-dimensional space Reduced two-dimensional space

x3
Figure 12.4 The working mechanism of principal component analysis

Algorithm: Principal component analysis
Input: M ¼ fm1 ; m2 ; m3 ; . . . ; mN g and d

PN
Step 1: Calculate the mean m ~ ¼ N1 mj
j¼1 N
P T
Step 2: Calculate the covariance matrix CovðmÞ ¼ N1 ~ j mj m
mj m ~j
j¼1
Step 3: Decompose the spectral of Cov(m) generating Eigenvectors V1, V2, . . . ,
VD and corresponding Eigenvalues E1 ; E2 ; . . . ; ED sorted in descending order
E1 E 2 . . . ED 0
Step 4: For any mRD , the output with the reduced dimension will be
T
Y ¼ V1T ðm m ~ Þ; V2T ðm m
~ Þ; . . . ; VdT ðm m~Þ RD
and the approximated value of original m may be denoted by

mm ~ þ V1T ðm m~ Þ V1 þ V2T ðm m~ Þ V2 þ . . . þ VdT ðm m
~ Þ Vd
12.3.4 Classification using support vector machine

Classification plays a vital role in understanding human behavior and emotions.
SVM is one of the popular methods applied in emotion classification [48]. It is often
used to classify human emotional states [49], that are captured by EEG [50,51]. In
maintaining the accuracy and execution time trade-off, we can trust hierarchical
SVM classifiers [52]. SVM in existing literature is also used to classify context-based
music emotion recognition and classification [53], speech emotion recognition [54],
privacy-preserving [55,56], vision based on paste detection [57], classification of
agricultural crops [58], hyperspectral remote sensing images [59], pattern classifi-
cation [60,61], identification of brain structure [62], hybrid image denoising [63],
gesture identification in sign language applications [64], hand gesture recognition,
gesture recognition during dance [65], tweet act classification [66,67], etc.
SVM creates a decision boundary between classes in multidimensional space
so that testing data points can be easily classified in the correct category according
to the requirement [68]. The best decision boundary is called the hyperplane, to
design which, some extreme data points in the form of vectors get selected and
called as support vector. It is a binary classifier that classifies data points into two
classes. In classification problems, where data points need to get classified in more
than two classes, SVM applies a one versus all or one versus rest approach which
splits the classification problem having multi-class into one binary classifier for
every class. SVM can be of two types, i.e., linear SVM and nonlinear SVM applied
to linearly separable and nonlinearly separable data, respectively.
Figure 12.5 shows the working mechanism of SVM as a classifier. In the
concerned two-dimensional space, two classes can be easily separated using
straight lines. But there may be more than one straight line that correctly separates
Optimal hyperplane
Support Maximized
vectors margin
(a)
Y
3D representation of
Y nonlinear data
X
Z
(b) X Best hyperplane
Figure 12.5 The general working mechanism of support vector machine as a

classifier: (a) linear support vector machine and (b) nonlinear
support vector machine
the classes as shown in the blue line. Hence SVM aims to create a decision
boundary, and the best decision boundary among several possible boundaries gets
called a hyperplane. The distances between the hyperplane and the support vectors
are called margins, and the hyperplane having maximum margin between support
vectors is called the optimal hyperplane. For nonlinear data, one more dimension z
has to be added since no single straight line can act as the boundary between the
classes and it can be calculated by
z ¼ x2 þ y2 : (12.6)
In three-dimensional space, the hyperplane looks like a parallel to the x-axis
whereas in two-dimensional space, considering z¼1, the circumference can be
obtained with radius as 1 unit.
12.4 Experimental outcomes and discussions

The experimental outcomes are described here, and the classification performance is
evaluated using some performance metrics, namely accuracy, precision, error rate, and
recall. Python 3.7 and MATLAB R2015A have been used to simulate the system.
12.4.1 Dataset preparation

A publicly available dataset, named GAMEEMO [69], has been used in our pro-
posed framework to process and analyze EEG signals for the detection of stress or
anxiety present in the subject’s mind. Four different android games, namely
Slender: The Arrival, Goat Simulator, Unravel, and Train Sim World, have been
played by 28 different subjects with a total duration of 20 min (each game for 5
min). The games are of horror, fun, cool, and boring genre, respectively. Fourteen
electrode channels were considered to collect the data, namely AF3, AF4, T8, T7,
P8, P7, O2, O1, FC6, F8, F7, F4, F3, and FC5.
12.4.2 sLORETA-based activated brain region selection

The selection of activated brain regions is performed by assessing electrical actions
of the intracortical circulation of the EEG signal collected during game playing
using the sLORETA 2.0 platform [70]. We identify that prefrontal, occipital, par-
ietal, and temporal cortices have the peak activation for the two mental states
considered by us, i.e., stressed and not stressed. sLORETA solutions of the right,
left, back, front, bottom, and top views for the two mental states: stressed and not
stressed are illustrated in Figures 12.6 and 12.7.
12.4.3 Discrete wavelet transform–based feature extraction

outcome
Based on the sLORETA activation result, in the proposed framework, features got
extracted from the brain’s prefrontal, frontal and temporal region using ten elec-
trode channels, i.e., AF3, AF4, O1, O2, P7, P8, FC5, FC6, T7, and T8 by DWT
resulting in 38 25210284-dimensional feature space for 28 different subjects
and 4 games. Haar has been considered the wavelet basis function in the proposed
approach since it has performed better than the sym5, db1, and coiflet basis func-
tion applied in the existing literature concerning EEG feature extraction.
Approximate and detailed coefficients obtained from one subject and one game-
L A R L S R A S P
Sloreta 1 P Top Sloreta 1 I Back Sloreta 1 I Left

R A L R S L P S A
Sloreta 1 P Bottom Sloreta 1 I Front Sloreta 1 I Right
Figure 12.6 sLORETA-based brain activation for the class: not stressed
L A R L S R A S P
Sloreta 1 P Top Sloreta 1 I Back Sloreta 1 I Left

R A L R S L P S A
Sloreta 1 P Bottom Sloreta 1 I Front Sloreta 1 I Right
Figure 12.7 sLORETA-based brain activation for the class: stressed
based EEG data for four wavelet functions, namely Haar, db1, sym5, and db5, have
been shown in Figure 12.8. From the results, it can be said that the Haar wavelet
function is the best choice for the proposed work.
12.4.4 Principal component analysis–based dimensionality

reduction outcome
For reducing the number of features leading to the enhancement of performance,
PCA has been applied as the dimensionality reduction tool considering seven
principal components and reducing the feature space from 38 25210284
dimensions into 38 2527284 dimensions. As it is not feasible to show a seven-
dimensional PCA outcome in a 2D space, only considering the two most relevant
principal components detected from extracted features by DWT for the
GAMEEMO dataset, Figure 12.9 is drawn.
12.4.5 Support vector machine–based classification

outcome
The mental state classification was performed using SVM. The user-independent
classification got trained using data from all participants. For two reasons, we chose
SVM over alternative methods for categorization. To begin with, it has good gen-
eralization and accuracy capability with a short training sample. It is the most
reliable classification method for real-world settings, and it has been used to solve
different types of problems, including facial recognition, handwriting character
identification, and intrusion detection. It is created as a binary classification algo-
rithm, but it has been improvised to multiclass problems.
We have used SVM for stress detection in our proposed framework. To train
the classifier, the instantaneous frequency value obtained for all five EEG cycles is
Approximation coefficients Detail coefficients Approximation coefficients Detail coefficients
Level 1
Level 1
0 100 200 300 0 100 200 300 0 100 200 300 0 100 200 300
Level 2
Level 2
0 100 200 300 0 100 200 300 0 100 200 300 0 100 200 300
Level 3
Level 3
0 100 200 300 0 100 200 300 0 100 200 300 0 100 200 300
Level 4
Level 4
0 100 200 300 0 100 200 300 0 100 200 300 0 100 200 300
Level 5
Level 5
0 100 200 300 0 100 200 300 0 100 200 300 0 100 200 300
(a) (b)
Approximation coefficients Detail coefficients Approximation coefficients Detail coefficients
Level 1
Level 1
0 100 200 300 0 100 200 300 0 100 200 300 0 100 200 300
Level 2
Level 2
0 100 200 300 0 100 200 300 0 100 200 300 0 100 200 300
Level 3
Level 3
0 100 200 300 0 100 200 300 0 100 200 300 0 100 200 300
Level 4
Level 4
0 100 200 300 0 100 200 300 0 100 200 300 0 100 200 300
Level 5
Level 5
0 100 200 300 0 100 200 300 0 100 200 300 0 100 200 300
(c) (d)
Figure 12.8 Approximation and detailed coefficients (for five levels) applied on
GAMEEMO dataset [69]: (a) Haar function, (b) sym5 function, (c)
db5 function, and (d) db1 function
2 component PCA
Stressed
12.5 Not stressed
10.0
Principal component 2
7.5
5.0
2.5
0.0
–2.5
–5.0
–2.5 0.0 2.5 5.0 7.5 10.0 12.5 15.0
Principal component 1
Figure 12.9 PCA-based dimensionality reduction outcome

employed as a feature vector. The game genres described in [69] have influenced us
to classify the data points into two labels: stressed and not stressed. The players
playing the cool and funny games are labeled as not stressed, and the players
playing the horror and boring game have been labeled as stressed during the
training process.
12.4.6 Performance metrics

The most commonly used performance metric accuracy can be defined by the ratio
between the instances properly classified and the total existing instances in the whole
system. Precision can be defined by the ratio between properly predicted positive
remarks to the total predicted positive remarks. Recall also called sensitivity can be
denoted by the ratio of correctly predicted positive remarks to all remarks in the actual
positive class. The specificity of a classifier is the ratio between how many samples
are correctly classified as negative to how many samples are negative. The error rate
can be defined by the ratio between the total number of instances wrongly classified
and the total instances in the whole system. F1 score can be defined by the weighted
mean of precision and recall. The false-positive rate can be denoted by
false positive
False positive rate ¼ (12.7)
false positive þ true negative
and the true positive rate can be denoted by
true positive
True positve rate ¼ : (12.8)
true positive þ false negative
For the comparative study, other classifiers, namely kNN, DT, random forest
(RF), and naı̈ve Bayes (NB), have been compared with our proposed SVM in terms
of accuracy. For SVM, four types of kernel parameter values have been considered
here, i.e., linear, sigmoid, RBF, and polynomial, among which, RBF has out-
performed the others during the hyperparameter tuning phase.
Apart from the abovementioned performance metrics, we have also taken into
consideration some other parameters like receiver-operating characteristic (ROC)
which is a proof of identity of the classifier system [71] and the discrimination
threshold has been varied with the binary classifier’s parameter changes. In our
approach, along with ROC, we have also considered area under the curve (AUC)
values.
12.4.7 Performance evaluation

Performance metrics as described in Section 12.4.6 obtained from the classification
result of SVM are depicted in Table 12.2. The other classifier’s performance compar-
ison with SVM based on accuracy is illustrated in Figure 12.10 from where it is obvious
that the proposed SVM is more promising to detect the stress in a subject during playing
four different games. Figure 12.11 shows ROC curves, and AUC values obtained from
those four ROC curves are shown in Table 12.3. From the results, it can be said that
Table 12.2 SVM-based classification outcome
Mental state Precision Recall Error F1 score

Not stressed 0.88 0.63 0.08 0.69
Stressed 0.78 0.87 0.07 0.74
Avg. 0.83 0.795 0.075 1.43
100 90.71 87.58 80.36 74.02 92.79

90
80
Accuracy in percentage
70
60
50
40
30
20
10
0
kNN DT RF NB SVM
Figure 12.10 Comparison of proposed SVM classifier with other classifiers
Receiver operating characteristic (ROC) curve Receiver operating characteristic (ROC) curve
1.0 ROC 1.0 ROC
0.8 0.8
True positive rate
True positive rate
0.6 0.6
0.4 0.4
0.2 0.2
0.0 0.0
0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0
(a) (b) False positive rate
False positive rate
Receiver operating characteristic (ROC) curve Receiver operating characteristic (ROC) curve
1.0 ROC 1.0 ROC
0.8 0.8
True positive rate
True positive rate
0.6 0.6
0.4 0.4
0.2 0.2
0.0 0.0
0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0
False positive rate False positive rate
(c) (d)
Figure 12.11 Receiver-operating characteristic curve for the four kernels of SVM:
(a) for linear, (b) for sigmoid, (c) for RBF, and (d) for polynomial
RBF kernel-based SVM having a gamma value of 0.01 (marked as bold) is the best
choice for the current research work [72].
The comparison of our SVM classification result has been carried out by
considering accuracy as the metric with the existing related works [29,31,32] in
Figure 12.12. From the evaluation of effectiveness and applicability [73] of the
proposed approach, it is seen that the proposed framework works better than the
related ones.
12.4.8 Statistical significance using t-test

Table 12.4 depicts the mean and standard deviation values mentioned within the
bracket for each of the performance metrics by running five different algorithms 25
times. The means obtained from each metric from the best and second-best algorithms
Table 12.3 Parametric sensitivity analysis of SVM as a classifier
Parameter name Parameter value Gamma value FPR TPR AUC

Kernel Linear – 0.512 0.616 0.569
Sigmoid – 0.498 0.543 0.501
Radial basis function 1 0.342 0.578 0.657
0.1 0.231 0.763 0.564
0.01 0.152 0.912 0.89
0.001 0.165 0.769 0.745
Polynomial – 0.340 0.712 0.672
100
92.79 90.71
90 87.58
80
74.02
70
Accuracy in percentage
60
50
40
30
20
10
0
0.5 1 1.5 2 2.5 3 3.5 4 4.5
SVM (29) (31) (32)
Figure 12.12 Comparison with state-of-the-art literature

Table 12.4 Comparison of different game emotion recognition algorithms for

25 trials
Performance metric SVM kNN DT RF NB Statisti-

cal sig-
nificance
Accuracy 0.93 0.90 0.87 0.80 0.74 þ
(0.161) (0.531) (0.657) (0.645) (0.314)
Precision 0.92 0.87 0.89 0.87 0.75 þ
(0.134) (0.634) (0.453) (0.564) (0.324)
Recall 0.93 0.45 0.67 0.78 0.81 þ
(0.168) (0.324) (0.498) (0.654) (0.134)
F1 score 0.86 0.74 0.67 0.78 0.74 þ
(0.013) (0.125) (0.054) (0.435) (0.175)
Average error rate 0.07 0.40 0.25 0.65 0.19 þ
(0.019) (0.232) (0.143) (0.435) (0.023)
have been presented also using the paired t-test. The statistical significance level of the
difference of the means of the best two algorithms have been presented in Table 12.4’s
last column, where “þ” represents the t-value of 49 degrees of freedom that is statis-
tically significant at a 0.05 level of significance by two-tailed test. The sample size for
all of the t-tests is set as 25. The best ones have been marked with bold.
Mental state detection using EEG and ML techniques is a rapidly emerging field in
this era. Stress affects the mental and physical health of people, and in the modern
era, it is a common problem for people belonging to different age groups. To get
valid results in detecting human mental pressure during the pandemic era, we need
reliable identification of human stress. To identify stress levels, channel selection is
also important. The proposed methods are as follows, collection of human brain
signals using EEG, feature extraction by DWT, dimensionality reduction by PCA,
and the classification by SVM. The current work has classified the mental state of the
gamer as not stressed and stressed with an accuracy of 92.79%. Brain maps have also
been generated to detect stress levels in the brain’s active regions.
The work can be further extended by considering different stimuli apart from
the game, like movie watching, music listening, social media posts for detecting
stress, and possible rehabilitative measure during the pandemic.
Acknowledgment
The first author (i.e., Ahona Ghosh) is grateful to the university for providing the
AICTE Doctoral Fellowship, with appointment letter Ref. No.2.2.1/Regis./Appt.
(AG)/Ph.D(ADF)/2021 dated 01.02.2021. The third author (i.e., Sriparna Saha) is
grateful to the university for providing research seed money, File No. 9.6/Regis./
SD/Mn.(SS)/2019 dated 19.06.2019 and UGC Start-up Grant under the scheme of
Basic Scientific Research, File No. F.30-449/2018(BSR) dated 21.11.2019.
References
[1] Javed B., Sarwer A., Soto E.B. and Mashwani Z.U.R. ‘The coronavirus
(COVID-19) pandemic’s impact on mental health’. The International
Journal of Health Planning and Management. 2020; 35(5): 993–996.
[2] Xiong J., Lipsitz O., Nasri F., et al. ‘Impact of COVID-19 pandemic on
mental health in the general population: A systematic review’. Journal of
Affective Disorders. 2020; 277: 55–64.
[3] Willman M., Kobasa D. and Kindrachuk J. ‘A comparative analysis of fac-
tors influencing two outbreaks of Middle Eastern respiratory syndrome
(MERS) in Saudi Arabia and South Korea’. Viruses. 2019; 11(12): 1119.
[4] Hasöksüz M., Kiliç S. and Saraç F. ‘Coronaviruses and SARS-CoV-2’.
Turkish Journal of Medical Sciences. 2020; 50(SI-1): 549–556.
[5] Baize S., Pannetier D., Oestereich L., et al. ‘Emergence of Zaire Ebolavirus dis-
ease in Guinea’. New England Journal of Medicine. 2014; 371(15): 1418–1425.
[6] Buheji M., Jahrami H. and Dhahi A. ‘Minimising stress exposure during
pandemics similar to COVID-19’. International Journal of Psychology and
Behavioral Sciences. 2020; 10(1): 9–16.
[7] Elbay R.Y., Kurtulmuş A., Arpacıoğlu S. and Karadere E. ‘Depression,
anxiety, stress levels of physicians and associated factors in Covid-19 pan-
demics’. Psychiatry Research. 2020; 290: 113–130.
[8] Schoenfeld J. ‘The Invisible Risk: Pandemics and the Financial Markets.’ In
Tuck School of Business Working Paper No. 3567249; 2020.
[9] Dash S., Chakraborty C., Giri S.K., Pani S.K. and Frnda J. ‘BIFM: Big-data
driven intelligent forecasting model for COVID-19’. IEEE Access. 2021; 9:
97505–97517.
[10] Aslam B., Javed A.R., Chakraborty C., Nebhen J., Raqib S. and Rizwan M.
Computing. 2021: 1–17.
[11] Ravi V., Narasimhan H., Chakraborty C. and Pham T.D. ‘Deep learning
based meta-classifier approach for COVID-19 classification using CT scan
and chest X-ray images’. Multimedia Systems. 2021: 1–15.
[12] Chakraborty C., Banerjee A., Garg L. and Coelho Rodrigues J.J.P. Internet
of Medical Things for Smart Healthcare – COVID-19 Pandemic, Big data
series. Singapore: Springer Nature; 2020.
[13] Garg L., Chakraborty C., Mahmoudi S. and Sohmen V.S. Healthcare
Informatics for Fighting COVID-19 and Future Epidemics. EAI/Springer
Innovations in Communication and Computing. Cham: Springer Nature; 2021.
[14] Wada K., Shibata T., Musha T. and Kimura S. ‘Effects of robot therapy for
demented patients evaluated by EEG.’ In IEEE/RSJ International
Conference on Intelligent Robots and Systems. IEEE; 2005. pp. 1552–1557.
[15] Ali M., Mosa A.H., Al Machot F. and Kyamakya K. ‘EEG-based emotion
recognition approach for e-healthcare applications.’ In 2016 Eighth
International Conference on Ubiquitous and Future Networks (ICUFN);
2016. pp. 946–950.
[16] Mahato S. and Paul S. ‘Electroencephalogram (EEG) Signal Analysis for
Diagnosis of Major Depressive Disorder (MDD): A Review.’ In Nath V. and
Mandal J. (eds.) Nanoelectronics, Circuits and Communication Systems,
Lecture Notes in Electrical Engineering. Springer, Singapore, vol. 511;
2019. pp. 323–336.
[17] Mohammadi M., Al-Azab F., Raahemi B., et al. ‘Data mining EEG signals
in depression for their diagnostic value’. BMC Medical Informatics and
Decision Making. 2015; 15(1): 1–14.
[18] Liao S.C., Wu C.T., Huang H.C., Cheng W.T. and Liu Y.H. ‘Major
depression detection from EEG signals using kernel Eigen-filter Bank
common spatial patterns’. Sensors. 2017; 17(6): 1385.
[19] Niemiec A.J. and Lithgow B.J. ‘Alpha-band characteristics in EEG spectrum
indicate reliability of frontal brain asymmetry measures in diagnosis of
depression.’ In IEEE Engineering in Medicine and Biology 27th Annual
Conference, Shanghai, China, 2005. IEEE; 2005. pp. 7517–7520.
[20] Cai H., Sha X., Han X., Wei S. and Hu S. ‘A pervasive approach to EEG-
based depression detection’. Complexity. 2018; 2018: 1–13.
[21] Mahato S. and Paul S. ‘Detection of major depressive disorder using linear
and non-linear features from EEG signals’. Microsystem Technologies.
2019; 25(3): 1065–1076.
[22] Lun X., Yu Z., Chen T., Wang F. and Hou Y. ‘A simplified CNN classifi-
cation method for MI-EEG via the electrode pairs signals’. Frontiers in
Human Neuroscience. 2020; 14.
[23] Fong S.J., Dey N. and Chaki J. Artificial Intelligence for Coronavirus
Outbreak. 1st edn. Springer, Singapore, vol. 74; 2020. p. XI.
[24] Ghosh A., Roy S., Mondal H., Biswas S. and Bose R. ‘Mathematical mod-
elling for decision making of lockdown during COVID-19’. Applied
Intelligence. 2021: 1–17.
[25] Dutta P.K., Ghosh A., De P. and Soltani M. ‘A proposed model of a semi-
automated sensor actuator resposcopy analyzer for ’covid-19’ patients for
respiratory distress detection.’ In 11th International Conference on Cloud
Computing, Data Science & Engineering (Confluence), India, 2021. IEEE;
2021. pp. 618–623.
[26] Adhikary S., Ghosh A., Das P., Roy S. and Bose R. ‘Scope of sentiment
analysis on news articles regarding stock market and GDP in struggling
economic condition’. International Journal of Emerging Trends in
Engineering Research. 2020; 8(7).
[27] Sagayam K.M., Andrushia A.D., Ghosh A., Deperlioglu O. and Elngar
A.A. ‘Recognition of hand gesture image using deep convolutional
neural network’. International Journal of Image and Graphics. 2021:
2140008.
[28] Datta P.K., Biswas A., Ghosh A. and Chaudhury N. ‘Creation of image
segmentation classifiers for sign language processing for deaf and dumb.’ In
2020 8th International Conference on Reliability, Infocom Technologies and
Optimization (Trends and Future Directions) (ICRITO). IEEE; 2020.
pp. 772–775.
[29] Ghosh L., Saha S. and Konar A. ‘Decoding emotional changes of android-
gamers using a fused Type-2 fuzzy deep neural network’. Computers in
Human Behavior. 2021; 116: 106640.
[30] Hajian-Tilaki K. ‘Receiver operating characteristic (ROC) curve analysis for
medical diagnostic test evaluation’. Caspian Journal of Internal Medicine.
2013; 4(2): 627.
[31] Ghosh L., Saha S. and Konar A. ‘Bi-directional long short-term memory
model to analyze psychological effects on gamers’. Applied Soft Computing.
2020; 95: 106573.
[32] Asif A., Majid M. and Anwar S.M. ‘Human stress classification using EEG
signals in response to music tracks’. Computers in Biology and Medicine.
2019; 107: 182–196.
[33] Saeed S.M.U., Anwar S.M., Khalid H., Majid M. and Bagci U. ‘EEG based
classification of long-term stress using psychological labeling’. Sensors.
2020; 20(7): 1886.
[34] Ahn J.W., Ku Y. and Kim H.C. ‘A novel wearable EEG and ECG recording
system for stress assessment’. Sensors. 2019; 19(9): 1991.
[35] Kalas M.S. and Momin B.F. ‘Stress detection and reduction using EEG
signals.’ In 2016 International Conference on Electrical, Electronics, and
Optimization Techniques (ICEEOT). IEEE; 2016. pp. 471–475.
[36] Priya T.H., Mahalakshmi P., Naidu V.P.S. and Srinivas M. ‘Stress detection
from EEG using power ratio.’ In 2020 International Conference on
Emerging Trends in Information Technology and Engineering (IC-ETITE).
IEEE; 2020. pp. 1–6.
[37] Liao C.Y., Chen R.C. and Tai S.K. ‘Emotion stress detection using EEG
signal and deep learning technologies.’ In 2018 IEEE International
Conference on Applied System Invention (ICASI). IEEE; 2018. pp. 90–93.
[38] Sharma R. and Chopra K. ‘EEG signal analysis and detection of stress using
classification techniques’. Journal of Information and Optimization
Sciences. 2020; 41(1): 229–238.
[39] Peng H., Hu B., Zheng F., et al. ‘A method of identifying chronic stress by
EEG’. Personal and Ubiquitous Computing. 2013; 17(7): 1341–1347.
[40] Virmani D., Minocha A., Goyal L., Malhotra M. and Gupta M. ‘Se. Re. Ne.:
Stress detection using EEG and ECG.’ In Proceedings of 3rd International
Conference on Computing Informatics and Networks: ICCIN 2020. Springer
Singapore; 2021. pp. 189–198.
[41] Al-Shargie F.M., Tang T.B., Badruddin N. and Kiguchi M. ‘Mental stress
quantification using EEG signals.’ In International Conference for
Innovation in Biomedical Engineering and Life Sciences. Springer,
Singapore; 2015. pp. 15–19.
[42] Al-Fahoum A.S. and Al-Fraihat A.A. ‘Methods of EEG signal features
extraction using linear analysis in frequency and time-frequency domains’.
International Scholarly Research Notices, 2014.
[43] Evans J.R. and Abarbanel A. (eds.). Introduction to Quantitative EEG and
Neurofeedback. London: Elsevier; 1999. p. XVIII.
[44] Teplan M. ‘Fundamentals of EEG measurement’. Measurement Science
Review. 2002; 2(2): 1–11.
[45] Al-Haj A. ‘Combined DWT-DCT digital image watermarking’. Journal of
Computer Science. 2007; 3(9): 740–746.
[46] Wu Y.L., Agrawal D. and El Abbadi A. ‘A comparison of DFT and DWT
based similarity search in time-series databases.’ In Proceedings of the Ninth
International Conference on Information and Knowledge Management;
2000. pp. 488–495.
[47] Turnip A. and Siahaan M. ‘Adaptive principal component analysis based
recursive least squares for artifact removal of EEG signals’. Advanced
Science Letters. 2014; 20(10–11): 2034–2037.
[48] Yang N., Yuan J., Zhou Y., et al. ‘Enhanced multiclass SVM with thresh-
olding fusion for speech-based emotion classification’. International Journal
of Speech Technology. 2017; 20: 27–41.
[49] Nguyen T., Bass I., Li M. and Sethi I.K. ‘Investigation of combining SVM
and decision tree for emotion classification.’ In Seventh IEEE International
Symposium on Multimedia (ISM’05), Irvine, CA, USA; 2005. p. 5.
[50] Mehmood R.M. and Lee H.J. ‘Emotion classification of EEG brain signal
using SVM and KNN.’ In 2015 IEEE International Conference on
Multimedia & Expo Workshops (ICMEW); 2015. pp. 1–5.
[51] Guo K., Candra H., Yu H., Li H., Nguyen H.T. and Su S.W. ‘EEG-based
emotion classification using innovative features and combined SVM and
HMM classifier.’ In 2017 39th Annual International Conference of the IEEE
Engineering in Medicine and Biology Society (EMBC); 2017. pp. 489–492.
[52] Datta S., Sen D. and Balasubramanian R. ‘Integrating geometric and textural
features for facial emotion classification using SVM frameworks.’ In Raman
B., Kumar S., Roy P., Sen D. (eds.) Proceedings of International Conference
on Computer Vision and Image Processing. Advances in Intelligent Systems
and Computing, Springer, Singapore; 2017. vol. 459, pp. 619–628.
[53] Han B.J., Rho S., Jun S., et al. ‘Music emotion classification and context-
based music recommendation’. Multimedia Tools and Applications. 2010;
47: 433–460.
[54] Yi-Lin L. and Gang W. ‘Speech emotion recognition based on HMM and
SVM.’ In International Conference on Machine Learning and Cybernetics;
2005. vol. 8, pp. 4898–4901.
[55] Vaidya J., Yu H. and Jiang X. ‘Privacy-preserving SVM classification’.
Knowledge Information Systems. 2008; 14: 161–178.
[56] Yu H., Vaidya J. and Jiang X. ‘Privacy-Preserving SVM Classification on
Vertically Partitioned Data.’ In Ng W.K., Kitsuregawa M., Li J. and Chang
K. (eds.) Advances in Knowledge Discovery and Data Mining. PAKDD
2006. Lecture Notes in Computer Science. Springer, Berlin, Heidelberg,

vol. 3918; 2006.
[57] Ebrahimi M.A., Khoshtaghaza M.H., Minaei S. and Jamshidi B., ‘Vision-
based pest detection based on SVM classification method’. Computers and
Electronics in Agriculture. 2017; 137: 52–59.
[58] Foody G.M. and Mathur A. ‘Toward intelligent training of supervised image
classifications: Directing training data acquisition for SVM classification’.
Remote Sensing of Environment. 2004; 93(1–2): 107–117.
[59] Bazi Y. and Melgani F. ‘Toward an optimal SVM classification system for
hyperspectral remote sensing images’. IEEE Transactions on Geoscience
and Remote Sensing. 2006; 44(11): 3374–3385.
[60] Niaf É., Flamary R., Lartizien C. and Canu S. ‘Handling uncertainties in
SVM classification.’ In 2011 IEEE Statistical Signal Processing Workshop
(SSP); 2011. pp. 757–760.
[61] Chakraborty C., Gupta B. and Ghosh S.K. ‘Chronic wound characterization
using Bayesian classifier under telemedicine framework’. International
Journal of E-Health and Medical Communications. 2016; 7(1): 78–96.
10.4018/IJEHMC.2016010105.
[62] Akselrod-Ballin A., Galun M., Gomori M.J., Basri R. and Brandt A. ‘Atlas
Guided Identification of Brain Structures by Combining 3D Segmentation
and SVM Classification.’ In Larsen R., Nielsen M. and Sporring J. (eds.)
Medical Image Computing and Computer-Assisted Intervention – MICCAI
2006. Lecture Notes in Computer Science. Springer, Berlin, Heidelberg,
vol. 4191; 2006.
[63] Routray S., Ray A.K., Mishra C. and Palai G. ‘Efficient hybrid image
denoising scheme based on SVM classification’. Optik. 2018; 157: 503–511.
[64] Saha S., Bhattacharya S. and Konar A. ‘A Novel Approach to Gesture
Recognition in Sign Language Applications Using AVL Tree and SVM.’ In
Sa P., Sahoo M., Murugappan M., Wu Y. and Majhi B. (eds.) Progress in
Intelligent Computing Techniques: Theory, Practice, and Applications.
Advances in Intelligent Systems and Computing. Springer, Singapore,
vol. 518; 2018.
[65] Saha S., Konar A. and Roy J. ‘Single Person Hand Gesture Recognition
Using Support Vector Machine.’ In Maharatna K., Dalapati G., Banerjee P.,
Mallick A. and Mukherjee M. (eds.) Computational Advancement in
Communication Circuits and Systems. Lecture Notes in Electrical
Engineering. Springer, New Delhi, vol. 335; 2015.
[66] Saha S., Ghosh A., Konar A. and Nagar A.K. ‘Gesture recognition from
Indian classical dance using Kinect sensor.’ In 2013 Fifth International
Conference on Computational Intelligence, Communication Systems and
Networks; 2013. pp. 3–8.
[67] Chakraborty C., Gupta B., Ghosh S.K., Das D. and Chakraborty C.,
‘Telemedicine supported chronic wound tissue prediction using different
classification approach’. Journal of Medical Systems. 2016; 40(3): 1–12.
10.1007/s10916-015-0424-y.
[68] Noble W.S. ‘What is a support vector machine?’. Nature Biotechnology.

2006; 24(12): 1565–1567.
[69] Alakus T.B., Gonen M. and Turkoglu I. ‘Database for an emotion recogni-
tion system based on EEG signals and various computer games–
GAMEEMO’. Biomedical Signal Processing and Control. 2020; 60
(101951).
[70] Jatoi M.A., Kamel N., Malik A.S. and Faye I. ‘EEG based brain source
localization comparison of sLORETA and eLORETA’. Australasian
Physical & Engineering Sciences in Medicine. 2014; 37(4): 713–721.
[71] Kharshiing K.D., Kashyap D., Gupta K., et al. ‘Quality of life in the COVID-
19 pandemic in India: Exploring the role of individual and group variables’.
Community Mental Health Journal. 2021; 57(1): 70–78.
[72] Hussain M., Wajid S.K., Elzaart A. and Berbar M. ‘A comparison of SVM
kernel functions for breast cancer detection.’ In 2011 Eighth International
Conference Computer Graphics, Imaging and Visualization. IEEE; 2011.
pp. 145–150.
[73] Ranawana R. and Palade V. ‘Optimized precision-a new measure for clas-
sifier performance evaluation.’ In 2006 IEEE International Conference on
Evolutionary Computation. IEEE; 2006. pp. 2254–2261.
Chapter 13
Arduino-based robot for purification of
COVID-19 using far UVC light
C.N. Sujatha1, B. Sri Charan1 and K. Himabindu1
Abstract
The proposed work elaborates the necessity of sterilisation especially during
COVID-19 period. In this chapter, we will discuss about the development and
construction of an ‘Arduino-based robot for the purification of COVID-19 using far
UVC light’. Throughout history, the outgrowing technology has solved major
problems. Technology has made everything possible, from making knowledge
more available to more people and bringing group of people associated, to making
our planet cleaner and even saving lives. But the current circumstance of COVID-
19 is threat to human lives as there are high chances of getting sick in no time, so in
order to fight with this novel virus. There is necessity to keep our surroundings
clean and tidy. Cleanliness is the most important factor to keep ourselves away
from diseases. The objective of the current work is to discuss various inventions
made during the pandemic scenario and to contribute to the battle against COVID-
19 propagation, a novel human corona virus in hospitals, public transports, airlines,
and any enclosed areas with an approach, which is a basically a bot named
‘Arduino-based robot’ which uses ultraviolet (UV) sterilisation method to kill and
eliminate all the germs and viruses present in the surroundings and sanitises the
entire surroundings by cleaning and disinfecting the environment. Far-UVC
(ultraviolet C) radiation has a wavelength range of 207–222 nm. The decontami-
nating efficiency of far UVC light was previously tested by exposing bacteria that
were irradiated on a surface or in suspension. We have developed a UV-based
sterilisation method that uses single-wavelength far-UVC light produced by filtered
exclaims to selectively inactivate microorganisms while causing no biological
damage to exposed cells and tissues. The method is based on biophysical princi-
ples, as far-UVC light can penetrate and thus inactivate bacteria and viruses with
dimensions in micrometre or smaller, while far-UVC light cannot penetrate even
the outer dead-cell layers of human skin, nor the outer tear layer of an eye, due to
its high absorbance in biological materials. The far-UVC lamp plays a major role in
1
Department of Electronics and Communication Engineering, Sreenidhi Institute of Science and
Technology, Hyderabad, India
a disinfection robot. This bot will disinfect the virus and clean the areas; it will
travel autonomously and sterilise the areas without the need for human interven-
tion. In the discussion section, a demonstration of the bot’s implementation and
presentation was shown. This robot can also be used as a vacuum cleaner, elim-
inating the need for and expense of cleaning by humans.
Keywords: COVID-19; Sterilisation; Far-UVC lamp; Diseases; Bacteria;

UV and UVC lamp; UV sterilisation
13.1 Introduction
This chapter gives information about an Arduino-based far UVC (ultraviolet C)
sterilisation disinfection robot. It is a robot made by using ultrasonic sensor, far-
UVC lamp, accelerometer sensor, Arduino Uno. Arduino IDE (integrated devel-
opment environment) is the software used in building this robot for writing the code
that is been dumped in the Arduino Uno. The main purpose of this robot is to
sanitise surroundings and to make them free from harmful germs by using UV
(ultraviolet) rays. UV sterilisation is the method used to kill the microorganisms
using UV lamps [1]. In present days, UV, UVC, and far-UVC lamp are being used
for sterilisation process. The UV rays and UVC lamp may affect the tissues in
human body, so it cannot be used for sterilisation of surroundings in public places
and households. Under extreme conditions, they can be used in public places only
in the absence of all living organisms. The better way to solve the previous issue is
using a far-UVC lamp. Far-UVC lamps will not affect human tissues and give more
powerful and effective results when compared to the other lamps. Far-UVC lamps
make use of tiny wavelength far-UV light to eradicate or inactivate microorganisms
such as germs, bacteria, and viruses by exterminating their nucleic acids and dis-
rupting their DNA (deoxyribonucleic acid), preventing them from multiplying in
large numbers and rendering them incapable of performing essential cellular
functions. Our robot makes use of the far-UV excimer lamps [2]. This method is
based on the principle that very little penetration from the far UV lamp can cover
and kills the germs and viruses, and most importantly the far-UV light can neither
penetrate through the outer dead cell layers of human skin nor the outer surface of
the human eye. So, the far-UV light does not cause any harm to humans [3]. We
can also use few robotic vacuum cleaner components to make its function as a
robotic vacuum cleaner. With single robot, we can perform both cleaning and
sterilisation of surroundings.
13.1.1 Arduino
Arduino is an open-source platform which enables the users to use in software and
hardware projects easily. Arduino project began at Interaction Design Institute
Ivrea, Ivrea, Italy in the year 2005. The major goal of Arduino was to give a low-
cost and simple solution for students from all disciplines to create projects using
Arduino-based robot for purification of COVID-19 361
numerous sensors and actuators. The Arduino was named after the bar ‘Arduino of
Ivrea’, the place where the founders of the Arduino usually encountered.
Creative Commons attribution share-Alike 2.5 is responsible for hardware
designs and distribution of Arduino. Although the Arduino hardware and software
designs are accessible for free under copyleft licences, the developer has requested
that the name not be changed. ATmega8 AVR (Alf-Egil Bogen, Vegard Wollan,
RISC) microcontroller [4] is widely used in Arduino boards with various ranges of
flash memory, pins, and features. In 2012, 32-bit Arduino Due with Atmel
SAM3X8E was introduced. Single or double row pins or female headers are used to
make connections for programming the circuit as well as for incorporating into
other circuits. Arduino offers the capacity to link with ‘shields’, which are add-on
modules. For power supply, Arduino board manufactures mostly 5-V linear reg-
ulators and 16-MHz crystal oscillators, or ceramic resonators to generate clock
pulses. For the simplification of uploading programmes to on-chip flash memory,
Arduino microcontrollers are pre-programmed with a boot loader. Programmes are
loaded onto board via serial connections to another computer. But currently USBs
(universal serial buses) are being used to upload the programmes onto the board for
this purpose. USB-to-serial adapter chip-like FTDI (Future Technology Devices
International Limited) FT232 is being used. AVR chip with USB-to-serial firmware
is used in Arduino Uno model boards. It can re-programme itself by using its own
ICSP (in-circuit serial programming) header. Other boards, such as the Arduino
mini, rely on a separate USB-to-serial adaptor or Bluetooth.
Arduino board microcontroller has several I/O pins which help in connecting
various other circuits. The number of pins will differ from one version to another.
Arduino Uno board has 13 digital I/O pins and six analogue pins to provide ana-
logue output in the range of 0–1,023. It has ATMEGA328P had microcontroller in
Arduino [5]. Along with these pins, there are several other pins as follows: Tx pin,
used for transmission of data; Rx pin, used to receive data; Vin pin, helps in pro-
viding input voltage; and 5 and 3.3-V pins are used to supply voltage to external
devices.
Analogue pins act as an interface in order to convert analogue data into digital
format. With the help of these pins the code written in human understandable
language (English language) is converted into a hex file using compiler and stored
in memory of microcontroller.
There are several versions of Arduino boards. Some of them are as follows:
● Arduino RS (recommended standard) 232 (male pins)

● Arduino Diecimila
● Arduino Duemilanove (rev 2009b)
● Arduino UNO R2
● Arduino UNO SMD (surface-mounted device) R3
● Arduino Leonardo
● Arduino micro (ATmega32U4)
● Arduino Pro Micro (ATmega32U4)
● Arduino Pro (no USB)
● Arduino Mega
● Arduino Nano (DIO-30 footprint)
● Arduino LilyPad 00 (rev 2007) – no USB
● Arduino robot
● Arduino Esplora
● Arduino Ethernet (AVRþW5100)
● Arduino Yun (AVRþAR9331)
● Arduino Due (ARM (advanced RISC machines/Acron RISC machine) Cortex-
M3 core)
For the functioning of Arduino hardware, one has to dump the code in Arduino
board. The code for Arduino board can be written in any of other programming
languages like Java, C, Cþþ with their suitable compilers. The Arduino-integrated
development environment (IDE) is a Java-based cross-platform development
environment that runs on Windows, Linux, and macOS. This software is licensed
under the GNU Lesser General Public License (LGPL) or the GNU General Public
License (GPL). This software is initiated from IDE for handling and writing of
languages. Many code editing features like text copying, cutting and pasting,
searching and replacing, highlighting syntax, automatic indenting, and brace
matching are included in this software. This software is easy to use because the user
can compile and upload programmes using one-click mechanisms. With special
rules of code structuring, Arduino IDE supports C and Cþþ. It has software library
which contains many commonly used inputs and functions. The application
AVRdude is used to convert executable code into a hexadecimal-encoded text file.
Arduino Pro IDE (alpha version) was released on 18 October 2019. Although
this version still uses command line interface (CLI), there are improvements like
autocompletion support, git integration, and more professional development
environment. Eclipse Theia Open Source IDE is a base for front-end of this
application. Some main features of alpha release are fully modernised and featured
environment. Three modes are classic (similar to classic Arduino IDE), dual, and
pro (file system view).
● New board manager
● New library manager
● Board list
● Basic auto completion (Arm targets only)
● Integration with Git
● Serial monitor
● Dark mode
Though many microcontrollers and microcontroller platforms like Parallax
BASIC stamp, MIT’s (Massachusetts Institute of Technology) Handy Board,
Phidgets, Netmedia’s BX-24 are available for physical computing, but most of us
prefer Arduino for its easy to use nature.
Code is written as a block of statements in Arduino IDE software dump it in
Arduino board with USB portal. Later the power supply is given to the board
through power sources like power bank, power jack, or supply voltage to vin pin.
The code writer can check the functioning of code using serial monitor. The fol-
lowing are the steps involved in writing code in Arduino IDE and dumping:
● Write the code

● Save the file
● Compile the code to find errors
● Uploading of code into a board
● Checking values in a serial monitor
While writing the code, read and write statements are used in order to make the
Arduino board to take input values and give output based on the code. DigitalRead
command helps in scanning values at the pins. Digital write command is used to
print the values. In void setup (), one has to declare the usage of pins. There are
three phases in Arduino programme such as declaring of variables, Void setup ()
and Void loop (). Sinking-0 is set when Arduino is receiving values and Sourcing-1
is set when giving current values out. In this, we used Arduino UNO board for
completing the task of microcontroller.
13.1.2 Far-UVC lamp

With a sudden outbreak of coronavirus, people are very keen to keep environment
clean and sterile by cleaning at regular interval of times. Chemical disinfectants are
used in many areas to clean the environment, but it can have negative effects on
humans and they are ineffective against the virus. In current time, there is no proper
sterilising equipment. At the moment, there is no safe method to eliminate virus
completely. Public areas like hospitals have to sterilised perfectly in order to stop
virus transmission. Generally, all of us use germicidal UVC (GUVC) light (254
nm) that is made by phosphor coating and fused quartz rather than a borosilicate
glass bulb to eliminate viruses. Due to its longer wavelength, UVC light can cause
damage to eye and skin. So, one has to use another lamp which is used for ster-
ilisation and it should not cause any effect to humans. A far-UVC light is a disin-
fection device which is safe for humans and can be used for sterilisation method. It
does not cause any harm to humans because its wavelength is too small to penetrate
human skin but can still inactivate virus. Far-UVC light has 99.9% sterilisation
efficiency and is safe for non-target organisms. A far-UVC light has 222-nm
wavelength and a frequency of 1.4E9 MHz at 539 kJ/mole [6]. Both GUVC light
and far-UVC light have strength to penetrate through cells and have energy to
disrupt them. This may cause damage to DNA. Thus, radiations of their light can be
used as disinfectant or for sterilisation process. A far-UVC light is safe and alter-
native for GUVC light. There have been two studies performed with far-UVC to
prove that it has the same efficiency as GUVC light. Columbia Medical Centre of
Research was the first to study the effectiveness of far-UVC. They tested far-UVC
light effect on two corona strains like HCoV (human corona virus)-229E (VR-740)
and HCoV-OC43 (VR-1558) using human lung cells infected with the virus. After
experiment, the results were like HCoV-OC43 and HCoV-229E were eliminated by
Figure 13.1 A 222-nm far-UV lamp
95% in 11 min, 99% in 16 min, and 99.9% inactivation in 25 min. They concluded
that virus is high sensitivity to a far-UVC light. It can also eliminate the SARS-
CoV-2 (severe acute respiratory syndrome-coronavirus-2) virus as it is similar to
HCoV-OC43 and HCoV-229E [7]. The other study was done by Hiroshima
University and included SARS-CoV-2 containing solutions. They placed 100-mL
solution containing the virus spread onto a 9-cm sterile polystyrene plate [8]. They
placed far-UVC lamp near to that plate. After 30 s of exposure to far-UVC radia-
tion, 99.7% of the SARS-CoV-2 was eliminated in this experiment.
Mainly, we use a 222-nm excimer lamp with a 120-V power supply and
operate at 150 W as shown in Figure 13.1. This device is priced at $1,000. It is
high-efficiency germicidal lamp and has strength to prevent regrowth of bacteria
[9]. Design of a robot and installation of it in-house determines to observe the
working performance.
The main intention of our work is to eradicate germs, bacteria, and viruses
from the surroundings by designing a robot and placing it in the house/hall for the
following purposes:
● To keep the surroundings and objects clean and tidy by sanitising them timely.
● To reduce human efforts and to make life easier.
● To protect humans from becoming victims of dreadful viruses and diseases.
The remaining part of this chapter is organised as follows. In Section 13.2, we
present a review of previous efforts relevant to the proposed work followed by a
description of an Arduino-based robot. Section 13.3 gives the working and imple-
mentation of a proposed robot. Section 13.4 elaborates the performance of the
designed robot and provides discussion on its usages. Section 13.5 provides the
conclusion of this chapter with future enhancements.
13.2 Literature survey

Human beings are all living in the world where harmful things or diseases come
without a notice, so they have to be prepared to eliminate these harmful germs,
bacteria, and viruses for a safe and healthy living. At the moment, the entire world
is experiencing a COVID-19 epidemic, in which hazardous viruses are ruining our
lives and also lowering the economic levels of a country. Doctors are keeping their
great efforts to find a cure and scientists are putting their heart and soul into the
invention of the vaccine at the earliest to bring us out of this pandemic situation
[10–12]. There are several inventions came into light in this pandemic situation.
According to the papers written by various authors, the following is the list of some
such inventions.
From the paper written by Petrovic, Nenad, and Ðorđe Kocić, there is ‘IoT
(Internet of Things)-based solution for indoor safety’. This system measures the
temperature through thermal camera and also checks the usage of mask and social
distancing through computer vision techniques. Previously, there are individual
system which performs the task of temperature checking, social distance, and mask
monitoring. But in this paper, they discussed about the IoT system which is a
combination of all three functions. They used the Python version of open CV for
writing algorithms of image detection. MQTT (message queuing telemetry trans-
port) is used to act as a bridge to send messages from sensors to servers about the
studied details. Whenever a person stands at the door, the infrared thermometer
which is attached to Arduino Uno board reads the temperature. When temperature
is greater than normal temperature, MQTT message with temperature and location
is sent to the security. Doors will be opened automatically for normal temperature
people and can proceed to the next stage checking, i.e. mask. The image captured
from camera is converted into grey scale image, and Haar cascade classifier is used
to check the position of the mask. If the mask is not placed properly then the MQTT
sends message to the security. When a person passes through the two checking
stages and enters the hall the camera images are again converted into grey scale
images and the Haar cascade classifier is applied to images. When more than one
person is appeared within the set, threshold distance message will be sent to
security. So, by using the previous system, one can analyse the temperature, mask,
and social distance [13].
The paper written by Swapnili Karmore says the information about ‘medical
diagnosis humanoid (MDH)’. This MDH is a humanoid that helps to diagnose
whether the person infected with COVID-19 or not. It is a humanoid which works
using artificial intelligence (AI) for data processing. They also help in navigating
people in hospital by using various sensors like IR (infrared) sensor, camera
module, and automatic navigation. First the person is directed to sanitisation stage
and then directed to diagnosis stage. At this stage, diagnosis is done by using E-
Health kit, CT (computed tomography) scan for chest, blood sample collection and
processed using AI. This project is a cost-effective and helps medical staff to test
without contact [14]. The paper of Herbert Wanga, Thobius Joseph, and Mauna
Belius Chuma’s gives information about ‘Designing the low-cost water tap’. One of
the precautions to prevent COVID-19 is to frequently wash hands. A low-cost

automatic water tap was designed which works by solar power and micro-
controllers. It can be of great use in public places and gatherings. An infrared
sensor is used to detect the presence of hand or object. When one places their hand
under the hand, the infrared sensor detects and gives signals to Arduino. They used
ULN2803 transistor to amplify the signal and to send enough current to the aqua-
rium pump. The aquarium pump releases for setting the duration of time. The
power requirement of the entire system is fulfilled by the solar panel. The previous
IoT-based solution for indoor safety uses Raspberry Pi but in the automatic water
tap, they used Arduino Uno as a microcontroller [15].
Gordana Lastovicka-Medin presented a report on a prototype gadget for
increasing awareness and monitoring. This device alerts the user by vibrating when
the user tries to touches the face. According to this paper, it is studied that a person
touches his face for about 24 times in 1 h. But during this pandemic, touching of
face is not acceptable. So, the device built by them helps to avoid touching of face.
This device is in the form of glasses with ultrasonic sensors and servo motors. The
servo motor helps in the rotation of sensor in 180 degrees. A buzzer, connected to
Arduino board gives a signal when a person tries to touch his face [16]. The most
important precaution during this pandemic is sanitisation. It is necessary to sanitise
a person or object whenever he is going into public places. ‘Temperature detection
and automatic sanitisation and disinfection tunnel – COVID 19’, the paper of Goda
Vasantharao and S. K. Arifunneesa discusses about system that checks the tem-
perature and sanitises the person. The system designed by them uses temperature
sensor like the one that is used IoT-based solution for indoor safety along with
buzzer instead MQTT interface to alert the security. When the temperature of the
person is normal, the system activates the air blower and spraying gun. They have
used PIR (passive infrared) sensor to notice the motion of person in a tunnel, unlike
the automatic water tap where the ultrasonic sensor was used [17].
A detailed report on the prototype ‘Anonymity preserving IoT-based COVID-
19 and other infectious disease contact tracing model’ is presented in [18]. In the
COVID period, it is very important for contact tracing which reduces the outbreak.
Tracing the people and taking measures may control the spreading of infections for
many people. There were many software models for contact tracing and had an
issue like privacy issues of storing contact details, so it is not used as much in
Western countries. The authors came up with hardware solution with RFID (radio
frequency identification), and they used blockchain concept to store the data which
leads to security, i.e., privacy for the stored information. They have implemented
the model in Remix IDE and noted their simulations and done their prototype
successful. The RFID is placed for humans, animals, and things and it is connected
to blockchain, whenever the infected person is near to them, a notification will be
sent. This project tries to control the spread of COVID-19.
In [19], the authors discussed about the ‘Panic buying situation during COVID-
19 global pandemic’. COVID-19 outbreak created huge impact on people and their
livelihood. Due to COVID-19, many people lost their lives and many scientists
mentioned importance of keeping homes clean and hygienic to stop the outbreak of
COVID. This made humans to think about cleanliness and they started to buy
masks, sanitisers, toilet papers, etc. These items have become their first choice to
buy as they are willing to keep their surroundings clean. These items caused a huge
business for many people and the cost of these items was increased 10 times the
original cost and failed to meet the required demand. People are panic about the
situation and urge to keep clean and sanitised surroundings.
Onur Dogan, Sanju Tiwari, M. A. Jabbar, and Shankru Guggari have done a
research work on AI/ML (machine learning) techniques used in outbreak of
COVID-19. They discussed about those techniques in brief and explained how they
were used in detection and extracted the data. Mainly CNN (convolutional neural
network), RF (random forest), ResNet (residual network), and SVM (super vector
machine) approaches are the most used AI/ML techniques against COVID-19. The
main objective of AI/ML techniques in COVID-19 are collection, prediction, and
diagnosis [20]. ‘A review on telemedicine-based WBAN (wireless body area net-
work) framework for patient monitoring’ is the paper written by Chinmay
Chakraborty, Bharat Gupta, and Soumya K. Ghosh. This paper is about the research
towards issues and challenges that were faced during the development of patient
monitoring in telemedicine. Because of the emerging developments in the inte-
gration of communications and microelectronics and embedded system technolo-
gies, there was a structural change in patient monitoring and health information
delivery systems. They used wireless body area network (WBAN) infrastructure to
be mounted on the people in hospitals. Various sensors are used to generate ECG
(electrocardiography), EMG (electromyography), EEG (electroencephalogram),
temperature, and many more. In this paper, they covered the key components of
WBAN in terms of information collection, data processing, data storage, and
patient monitoring in this post. It also discusses WBAN’s scalable architecture for
providing real-time patient health data. The QoS (quality of service) requirements
for integrating internal and exterior sensors into an autonomous system are defined
by this architecture [21].
The paper written by Muhammad et al. explained about ‘Supervised machine
learning model for prediction of COVID-19 infection using epidemiology data
test’. This paper discusses about building an ML model by using supervised
learning technique to detect the COVID-19 infection using an epidemiology
labelled data set. It also includes algorithms like decision tree logistic regression
and nail by us support with commission and artificial neural network (ANN). This
model takes chest X-rays images as a data set and pre-trained models’ inspectionV3
ResNet50 and inspection ResNetV2, an auto waitress net pre-trained model, give
the highest accuracy that is 98%. The ResNet pre-trained can be used for frontline
workers as it can detect the disease at early stage [22]. Whereas in [23] authors
have developed a robot for sterilisation process, they used UVC lamp and ozone
light to have double disinfection. They carried out by using MATLAB (Matrix
Laboratory), Arduino IDE, and ANNs. They tested the ANN in MATLAB and they
tested the behaviour of a robot in Proteus 8 software. They used three micro-
controllers, i.e., Arduino. First, Arduino with ANN will decide the location that to
be carried out for disinfection process; second, Arduino for displacement; third,
Read available Update display

raw values (every 50 ms) Flash display
Start
Yes
Finger No SPO2<
present? threshold?
No
Yes
DC and moving Calculate Average rolling

Beat detection average
average filters BPM and SPO2
Figure 13.2 Block diagram of a pulse oximeter
Arduino is for obstacle detection. In this way the disinfection process is carried out
in their model [23]. Megantoro et al. designed a model for disinfection process, it is
user controller robot. The disinfection process will be done by using sprayer. It has
key components like LoRa (long range) sender/receiver, Arduino Mega 2560, GPS
(global positioning system), camera unit, non-contact temperature, accelerometer,
and humidity temperatures. They have created one UI integration, where user can
see the video streaming, sensors value, etc.; through this, a user controls the robot.
The sprayer will spray the disinfection liquid to objects and places and thus dis-
infection process is done in this model [24].
From the paper of Metcalfe, Benjamin there is ‘A cost-effective pulse oximeter
designed in response to the COVID-19 pandemic’ which works with the help of
optical sensors. They designed it by using Arduino Nano microprocessor,
MAX3010x pulse oximeter breakout board, and an SSD1306-based display. It is a
low-cost optical pulse oximeter that comes with a finger clip. They made a finger
clip by using the 3D printing process. This oximeter is convenient for many hours
of use as it works on a 9-V battery. Figure 13.2 shows the working of pulse oxi-
meter in the form of block diagram [25].
Social distancing is a must and should be a habit that has to be followed during
this pandemic, but there will be times where people forget about it or accidentally
violate it. Basil Reji’s paper says about the ‘smart cap’, the device which helps in
maintaining the proper social distance. They designed a cap with a PIR sensor
which detects the presence of human beings within a radius of 1 m and alerts the
person by buzzer sound and also with a message. They used an SIM (subscriber
identification module) 808 (GPS–GSM module) module was used to track the
location of the person and to send the message when the temperature exceeds the
normal range. A face shield was also for additional protection in cases of mis-
placement of mask. When the person using this cap is tested COVID positive, it is
easy to locate the primary contacts because of the SIM 808 (GPS–GSM module).
They used the Arduino Uno as a microcontroller to control the sensors. The entire
system works on a 9-V battery with a switch [26]. ‘Flexible (portable) COVID-19
detecting device using Arduino and Sensors’ was discussed in the paper written by
Katha Roy. This a portable device which can be fixed on the mask or wrist. They
designed the device based on the principle that the air human exhale is moist,
humid in vapour form, and warmer than the surroundings. When person violates the
social distance and enters the zone of exhaled gas, this device alerts the person with
a buzzer sound. This device will be a great help during this pandemic as the disease
is airborne. It is also light in weight unlike the smart cap [27].
Sashmita Raghav’s paper discusses about the device ‘Suraksha’ which helps in
maintaining social distance. They made a headgear using a PIR sensor to detect the
human. They preferred using of this sensor because of their 120-degree view
instead of ultrasonic sensors with a 30-degree view. So, to cover entire 360 degrees,
they used three PIR sensors. NodeMCU (node micro controller unit) ESP32, 32-bit
microprocessor with on-board Wi-Fi and Bluetooth was used to act as an interface
between the buzzer, sensors and to establish connection with a mobile phone to
exchange data instead of MQTT and SIM 808 which were used in IoT-based indoor
safety and smart cap, respectively. In the case of violating social-distance
NodeMCU ESP32 sends the information about duration and closeness of contact
and also gives the buzzer sound to alert the user. Figure 13.3 describes about steps
involved in working of a Suraksha device [28].
Dr R. Dhaya’s paper ‘Deep Net Model for Detection of Covid-19 using
Radiographs based on ROC (Receiver Operating Characteristics) Analysis’ dis-
cusses about the CNN models that are developed to find the infected persons using
chest X-rays. Their study helps to overcome shortage of test kits due to which
spreading of infection is rapid. There are three CNN models ResNet50, ResNetV2,
and InceptionV3 out of which ResNet50 is the fastest [29]. The paper of Singh,
Sundaram, and Aditya Garg discusses about ‘Automatic Door Handle or Knob
Sanitiser’. This contraption is powered by an Arduino Nano, an infrared sensor, and
a servo motor. It automatically sanitises the door knob that is used in public places.
Whenever the IR senses the hand at the door knob, the servo motor pulls the handle
of sanitiser bottle to release the liquid to clean the surface of knob and hand [30].
The paper on ‘A Review On Smart Health System To Monitor People In
Covid-19 Quarantine’ of Mr Sushilkumar Thakare and Mahadeo Kokate discusses
Positive
Input image Batch Average Routing by

Convolution pooling Convolution
normalisation agreement
Negative
Figure 13.3 Block diagram of the proposed COVID capsule model

about quarantine system. This system is an integration of pulse sensor and breathe
sensor. This system is designed by using Arduino Pro mini, the smallest of Arduino
series and HC-06 Bluetooth module to establish connection to doctors and family.
It helps in monitoring the blood pressure, blood pH (potential of hydrogen), heart
rate, body temperature, respiratory rate of the patient who is in home quarantine. If
any of the important parameters of the patient falls below the limit, it alerts the
family members and doctors through email and twitter messages with the help of
Bluetooth module [31]. Bhagwant, Gavade Poonam, and V. V. Purohit discussed
about the device used to check the safety regulations at entrance by using IR sen-
sors, temperature sensors. They made a device using three IR sensors, temperature
sensor, pulse oximeter interfaced with Arduino Uno microcontroller. Two IR sen-
sors will be placed outside the gate and one sensor inside the gate. When person
approaches the gate, temperature was measured by using contactless IR thermo-
meter and the person is asked to place the finger on the pulse oximeter. If the
person has normal temperature and blood oxygen levels greater than 90%, then the
person is allowed into the hall. Unlike ‘IoT-based solution for indoor safety’ device
which measures the social distance inside the hall, this device allows person to
enter the mall only if the sufficient place to maintain is available inside the hall. So,
when the person enters the hall, the IR sensor placed inside the gate automatically
increases the count [32].
The article in ‘The Print’ discusses about ‘Bag valve masks’ that was a low-
cost ventilator invented by students of IIT – Hyderabad. These ventilators are made
within the cost of Rs. 5,000 which is 100 times less than the conventional machine.
These are at present at 3D powered range. At present these are hand powered and
can be further developed to be operated with a simple power source like car bat-
teries. They are of a great help in this pandemic period where there is a large
requirement of ventilators [33]. The paper ‘Far-UVC applications in healthcare’
has written by Parker Esswein gives idea about various healthcare devices using
far-UVC. This paper tells us about the importance of a far-UV lamp in disinfection
and various applications using far-UV in health centres. A far-UVC box sanitiser
helps in sanitising objects and far-UVC disinfection wand can be used to sanitise
small objects like mobile phones. In these two systems, human support is required
and it is limited to few objects. This project is similar to our object but with no
human assistance [34]. The paper ‘Far UV-C lights and fibre optics-induced
selective far-UVC treatment against COVID-19 for fatality-survival trade-off’ by
Imran Haider, Asad Ali, Tooba Arifeen, and Abdus Sami Hassan gives information
about the effectiveness of far-UVC lights in eradicating corona virus. This paper
gives an idea about the usage of far-UV light in public places and for the treatment
of humans. A far-UVC light with the help of optical fibre can treat a person without
damaging the skin [35].
Simply using a sanitiser and mask would be a great help in protecting us from
the viruses, but to sanitise our surroundings, there is no such product available. This
bot is a product which helps us sanitise our surroundings without any harmful
effects. UV is the electromagnetic radiation in the range of 100–400 nm [36] as
shown in Figure 13.4. We have three lamps: UV (254 nm), which is harmful to
Light
UVC UVB UVA
Wavelength(nm)
100 280 315 400 700
Ultraviolet Visible Infrared
Figure 13.4 Band of spectrum
(a) (b)
Figure 13.5 (a) Corona cleaner-source and (b) the UV blaster
human skin, UVC lamp which is also dangerous to the human skin and the radiation
received from that emitter in surfaces depends on the location where it is present
[37]. Far-UVC (222 nm) is the third lamp which is not harmful to human skin and
does not contain the traces of mercury [38]. Therefore, it is safe to use in this
project, and hence it can be used for sterilisation process.
Existing solutions are displayed in Figure 13.5. Figure 13.5(a) shows a picture
of a corona cleaner which uses UVC light to inactivate the RNA (ribonucleic acid)
of the viruses thereby not allowing the transfer of viruses and bacteria. This device
sanitises all the portable electronic gadgets through UVC lamps. It is portable and
easy to use [39]. Figure 13.5(b) shows the picture of the ‘the UV blaster’, which
again uses UVC light to sanitise virus-prone areas. It is used in hotels, restaurants,
airports, shopping malls, metros, and factories. This can be operated with a remote
using a mobile or laptop using a Wi-Fi link. It includes six UVC lamps, each with a
UVC power of 43 W and wavelengths of 254 nm for 360-degree lighting [40].
‘The disinfection period for a space of around 1212-ft2 size is around 10
min’, says the concern, adding that a 400-ft2 area can be sanitised in 30 min if the
device is placed in various locations throughout the room.
13.2.1 Improvements and requirements

Our idea satisfies both the previous existing system but using far-UVC lamps,
which is not harmful and is more effective than the previously mentioned products.
Our product sanitises the complete surroundings like our home, classes, and offices.
Also, our idea does not require any human efforts. It is designed to be completely
autonomous.
13.2.1.1 Hardware requirements

In order to build our robot, various hardware components were used. The lists of
various components used in this robot are as follows:
● Ultrasonic sensors
● Far-UVC lamps
● Arduino Uno
● Motor driver
The brief description of each of the following hardware components is given in
the following subsections.
Ultrasonic sensors
Ultrasonic sensors are the electronic devices which can measure the distance of an
object or target in making use of ultrasonic waves. It measures distance by
releasing ultrasonic sound waves then translating reflected sound wave into elec-
trical signals. It basically comprises two primary components as shown in
Figure 13.6.
● Transmitter: used for emitting sound using Piezoelectric Crystals;
● Receiver: receives the sound after it travels to and from the target or object.
To determine the distance between the ultrasonic sensor and the object, the
sensor considers the time taken by the transmitter and the receiver. We know
Figure 13.6 Ultrasonic sensor

speed ¼ distance/time. For calculating the distance, we take D ¼ 1/2T C, where

D is the distance, T is the time, and C is the speed of sound, i.e., 343 m/s. As the
sensor covers the distance twice, i.e., from transmitter to object and object to
receiver so, we will take 1/2 in the formula for calculating distance. Ultrasonic
sensors are used as proximity sensors. They are used in automatic self-parking and
obstacle detection systems, anti-collision systems, and in manufacturing industry.
Ultrasonic sensors have a wide number of applications in the field of robotics. They
can also be used as level sensors to monitor, detect, and identify [41]. They are
employed in the medical field to create images of inside organs, cancers, and
unborn infants.
IR sensors are commonly employed as proximity sensors for obstacle
detection, although they have a few flaws, such as low precision when compared
to ultrasonic sensors. Obstacle detection is more accurate using ultrasonic sen-
sors. They will work by generating sound waves and waiting for the sound to
reflect back. The distance will then be calculated based on the time it takes to
reach the object and the time it takes to reflect back [42]. In some of the sensors,
they use separate emitter and receiver, but in ultrasonic sensor both are com-
bined in one package and it can be used for many applications where size is at a
premium. A radar also works in the same way as an ultrasonic sensor, but most of
the people use ultrasonic sensors as it is readily available in the market and these
are available at low cost, and they can detect the objects more accurately than
radar. As a result, many people prefer to employ ultrasonic sensors instead
of radar.
Technical specifications of an ultrasonic sensor: power supply – þ5 V DC;
quiescent current – <2 mA; working current – 15 mA; effectual angle – <15
degrees; ranging distance – 2–400 cm/100 to 13 ft; resolution – 0.3 cm; and mea-
suring angle – 30 degrees.
Far-UVC lamps
A method for limiting or stopping viral transmissions in the air and inactivating
them in a short period of time is possible. Germicidal UV light would be effective,
but when directly used, would be hazardous to the human eye and skin. In contrast
to this, far-UV lamps would be effective as they emit far-UVC light which kills
pathogens and bacteria efficiently and inactivates their cellular functionality caus-
ing no harm to humans [43]. According to researches, it is said that on a continuous
exposure of far-UVC light for 25 min, 99.9% germs and bacteria would be eradi-
cated. Therefore, the far-UV lamps form the heart of our project.
Arduino Uno
The Arduino Uno is a microcontroller board which is based on ATmega328p [44].
It has 14 digital input/output pins out of which 6 pins are used as PWM outputs, 6
analogue inputs, a 16-MHz crystal oscillator, 4 hardware serial ports, a power jack,
an ICSP header, and a USB connection as shown in Figure 13.7. This board is
designed for low-cost projects. It has a flash memory of 32 kb, EEPROM (elec-
trically erasable programmable read-only memory) of 1 kb, and SRAM (static
random-access memory) of 2 kb.
Figure 13.7 Arduino Uno
Figure 13.8 Motor driver
Motor driver
Motor drivers form an interface between the motors and control circuits. It has
a voltage regulator, a power led, a 5-V jumper ENA, and ENB pins. IN1, IN2,
IN3, and IN4 are direction control pins at the edges of motor driver as shown in
Figure 13.8. They require a large amount of current whereas low current sig-
nals are worked on by the controller. So, the job of a motor driver is to trans-
form a low-current signal into a high-current signal that may be used to drive
a motor.
12-V battery
Lead–acid batteries are used in marine industries. In general, it has two 6-V bat-
teries in series, or a single 12-V battery. Batteries are arranged with several single
cells in series manner, where each cell produces approximately 2.1 V. In a 12-V
battery, six single cells are arranged in series producing an output voltage of 12.6
V. A battery is a mostly used component in projects. In the place of a battery, we
can use voltage regulators based on requirement of applications.
13.2.1.2 Software requirements

We need to develop the code for functioning of the bot. We need to dump the code
we developed in the Arduino for functioning of the bot. For this purpose, we use the
software called Arduino IDE. The Arduino-integrated development environment is
software which comprises a text editor in which code is to be written, a text con-
sole, a toolbar which has buttons to perform collective functions and a list of
menus. It connects to an Arduino and Genuino hardware for uploading programmes
and also sets up a mode of communication between them. This is the main text
editing programme used for Arduino Programming. First, the IDE interprets,
translates, and compiles the sketches into codes that Arduino can understand. After
compilation, the Arduino code is dumped into the board’s memory.
In Figure 13.9, the circuit diagram of the proposed Arduino-based far-UVC
robot is displayed. Figure 13.9 shows the design of the proposed idea using circuit.
io software. It is a software tool where we can stimulate and design our model. In
this software Arduino is interfaced with ultrasonic sensors and motor driver to get a
clear idea about the connections. On stimulating this design in this software, we can
know about the errors in our design and rectify them before going for the imple-
mentation of it in hardware. The ultrasonic sensors have four pins-Vcc, ground, Tx,
and rx pins. The Vcc pin is connected to 5-V Arduino. The Tx pin is connected to
digital pin, and rx pin to digital pin. The motor driver is connected to four pins of
Arduino. The gear motors will be connected to out pins of motor driver. The far-
UVC lamp will be connected to Arduino.
Figure 13.9 Component interfacing circuit diagram

13.3 Working of the proposed robot

First, we make a box out of wood. At the bottom surface of the box, four wheels
would be attached at the four corners for the robot to move. Four LED (light-
emitting diode) far-UVC lamps would be put in the centre of each edge, on the four
sides of the box. These lamps on emitting UV light would kill germs and bacteria
and thus sanitise or clean the entire floor by moving on wheels.
Four pole-shaped far-UVC lamps would be positioned at the four corners of
the box’s top surface, helping to sanitise the entire surroundings and objects around
us. To move back and forth, this bot does not require the use of a remote control or
an app. It moves on its own like a robotic vacuum cleaner and contains vacuum
cleaner components, so it acts as vacuum cleaner as well as a sterilisation robot.
Figure 13.10(a) is the block diagram which represents the various steps and devices
involved in the design of the bot. Figure 13.10(b) gives the information about step-
wise process involved in the function of the bot. From Figure 13.10(a) and (b), it is
Ultrasonic sensor
Arduino
Uno
Gear
Motor driver
motors
(a)
Start
Far UVC lamp on

Arduino
Motor driver
Ultrasonic sensor
detects the obstacle
Motors Motors Motors Motors
Wheels Wheels Wheels Wheels
(b) Locomotion
Figure 13.10 (a) Block diagram of the bot and (b) flow chart describing the
design flow
(a) (b)
Figure 13.11 (a) Front and (b) side view of the project
evident that the Arduino Uno is a key component of the proposed work. This is the
brain of the proposed work as it controls and sends the information; here the
ultrasonic sensor detects the objects and sends the information to Arduino and it
decides what operation has to be done. Based on ultrasonic observations, Arduino
Uno commands the motor driver to regulate the gear motors.
There are components like ultrasonic sensors which help the bot to detect
obstacles in their paths and move forward. This entire set-up can be employed in
houses, offices, malls, theatres, and hospitals. This set-up would sanitise the entire
place in about 15 min. It would cause no harm to humans and would prove to be a
boon to many as it is automatic and needs no human efforts to operate it and also
helps us keep our surroundings and objects clean, thereby protecting us from dan-
gerous diseases or viruses and keeping us clean, hygienic, and healthy. CAD
(computer-sided design) models of the proposed idea are shown in Figure 13.11.
Figure 13.11(a) and (b) shows the view of a CAD model of our project. The
two parts show the box in which ultrasonic sensors, Arduino Uno, and other elec-
trical components are placed internally. On the top of the surface, pole-shaped far-
UVC lamps are placed as shown in Figure 13.11 for sanitising the surroundings and
at the bottom surface, four LED shaped far-UVC lamps are placed for sanitising the
ground.
13.3.1 Value proposition

The prototype of our product might cost as much as Rs. 8,000. We intend to build
the robot ourselves to save money. The far-UVC lamp would cost between Rs.
3,000 and Rs. 4,000. However, the cost of mass manufacture of the far-UVC lamp
will be lower. The robotic vacuum cleaners have a high demand nowadays, as we
combining both cleaning and sanitising in our robot; there will be a high market
range for this product. It can also be utilised in hospitals, as hospitals are locations
where cleanliness is prioritised. As a result, this sort of solution will be beneficial in
sanitising the rooms, cleaning the floors, and reducing the efforts of staff. The
cleaning team is at greater risk during COVID; however, this model aids them in
the cleaning process and decreases the risk. It is very significant in the hospitality
industry. It can be used by all households, industries, offices, hospitals, and other
places to maintain the cleanliness.
13.4 Results and discussions

With the outbreak of COVID-19 pandemic, sanitisation of surroundings is given
most importance. But this task requires a significant amount of human effort. There
is also the possibility that the individual who sanitises our environment will become
sick with a virus. This risk is greatly decreased by the use of this bot.
We have done the prototype and placed in our home; it moves throughout the
entire house but does not cover the coroner areas. In balcony, it was moving
without any errors and showing accurate results. Ultrasonic sensor plays a major
role in detecting the objects, and it was spotting the objects as well as sending the
values to Arduino, and the locomotion is accurate in the use of ultrasonic sensor. As
we have four ultrasonic sensors, it detects the front, sides, and back perfectly and is
precise in detecting the objects, and the locomotion of the robot is accurate and
autonomous.
Figure 13.12 shows the interfacing of ultrasonic sensor and Arduino. The
ultrasonic sensor detects the objects and sends back the information to the Arduino.
We have kept ranges for the ultrasonic sensor by using speed–distance formula.
Distance ¼ speed time/2
Speed of sound is 343 m/s.
Based on the previous two conditions, we decide the range of an ultrasonic
sensor (depending on our requirement, we have to give time or distance values).
Figure 13.12 Arduino interfaced with ultrasonic sensors

Figure 13.13 Interfacing of Arduino, motor drivers, and gear motors
Interfacing of the Arduino with a motor driver and gear motors is shown in
Figure 13.13. Whenever the ultrasonic sensor detects the object, the information is
sent to the Arduino. Now Arduino determines and sends the instruction to motor
driver regarding the direction of movement. The motor driver regulates the power
to move gear motors.
The robot model was successfully created in AutoCAD, and the prototype was
completed as shown in Figure 13.14. The ultrasonic sensor is connected to Arduino
of 2, 3, 4, 5 pins, and other ultrasonic sensors were connected in the same way. The
transmitter pin will send the wave and when it identifies the object, it sends the
echo wave to receiver pin of ultrasonic sensor that goes to Arduino; now Arduino
decides and sends commands to a motor driver, then the motor driver controls the
gear motors. In this way, the robot locomotion works. The far-UVC lamp will be
ON while the robot is moving. The lamp will sanitise the room while it was moving
and at the same time it will clean the floor.
After implementing this robot steriliser using Arduino and ultrasonic sensors,
it has been observed that some difficulty aroused from ultrasonic sensor, like get-
ting trouble while it was moving in corner areas of rooms or halls. To overcome
such problems, LIDARs (light detection and ranging) can be used in the place of
ultrasonic sensors along with an accelerometer sensor in the future enhancement of
the proposed work. LIDAR and accelerometer sensors make the robot into a
complete autonomous. A typical remote-sensing method for estimating the exact
distance of an object on the earth’s surface is LIDAR. It benefits us by sending out
pulsed light waves into the environment. These pulses re-enter the sensor after
Figure 13.14 Prototype of an Arduino-based far-UVC robot
bouncing off nearby objects. The sensor calculates the distance travelled by using
the time it took for each pulse to return to the sensor. So that it can recognise
corners and walls in less time, allowing us to do the sanitisation in less time. In
addition to that, we can develop an app for the proposed work so that it would be
more helpful for the people in controlling the robot remotely. This bot will be a
great help in sanitising the surroundings and making them clean and tidy, by era-
dicating germs and preventing from being infected from dangerous virus. This bot
can be used in public places like malls, theatres, parks, and schools to reduce the
chance of persons being affected.
We can add a feature to this model like making the robot climbing up the walls
and cleaning the walls too. In order to achieve this, we need to change the shape of
the model such that it can climb the walls. Then the robot will be capable of
cleaning walls and floors. It would be great advantage for the people. The mobile
application can also be developed so that we can control it and if needed we can
change the option either it is being used for sanitising the floor or walls. In a
hospital the patient/staff can also use it if they needed for cleaning. The same robot
can be used for monitoring the patient condition in the hospital room using AI/ML
techniques. The robot will supply food to patients, it checks the temperature and
oxygen level of patients and sends data to store using blockchain technology, and
this feature reduces the work of staff and provides comfort for patient too.

In this pandemic and post pandemic situation, the proposed target will keep sur-
roundings clean by sanitising and continues the sanitisation process at regular
interval of times. Generally, as cleaning of a certain area or surface requires a lot of
human effort, this design is employed in order to decrease the human efforts and at
the same moment to have instantaneous cleaning. So, our proposal in this chapter is
very keen on this factor and made it to use less human effort by making it an
autonomous robot and sanitising the surroundings in short interval of time. The cost
is also very less which is affordable by all categories of people. Far-UVC lamps
have been used in sanitising the surroundings as they do not affect the humans. This
product when once released in market would prove to be a boon to the society as it
requires no human efforts and is completely automatic. Adding up to this, this robot
most importantly safeguards the health of humans by eliminating all the dreadful
germs and viruses. In the present COVID time, it is highly preferred for COVID
treat hospitals where sanitisation is very essential to protect the doctors, nurses,
serving people in the premises. Also, the far-UV light generated from the far-UVC
excimer lamps would cause no harm to humans and its cost is quite affordable. It
cleans the entire surroundings and objects in very less time and with no requirement
of human efforts. This bot would definitely hike up the market levels and would be
a great contribution to the society. In future, we can use a LIDAR sensor for
accurate automatic locomotion of a robot and an app to control the robot. We can
also incorporate features of robotic vacuum cleaners and design the robot to climb
the walls such that it cleans and sanitises the walls, floors, and other surfaces;
further this robot can be used in hospitals for supplying food, checking temperature
and oxygen levels of a patient. The patient and staff can control the robot using
mobile application which in turn makes hospitals into smart hospitals.
References
[1] Armani AM, Hurt DE, Hwang D, McCarthy MC, and Scholtz A. ‘Low-tech
solutions for the COVID-19 supply chain crisis’. Nature Reviews Materials.
2020;5(6):403–6.
[2] Welch D, Buonanno M, Grilj V et al. ‘Far-UVC light: A new tool to control
the spread of airborne-mediated microbial diseases’. Scientific Reports.
2018;8(1):01–07.
[3] Olcay A, Albayrak SB, Akturk IF et al. ‘A new Far-UVC based method for
germ free hospitals and travel: Initus-V’. medRxiv. 2021.
[4] Louis L. ‘Working principle of Arduino and using it as a tool for study and
research’. International Journal of Control, Automation, Communication,
and Systems. 2016;1(2):21–9.
[5] Badamasi YA. ‘The working principle of an Arduino’. 11th International
Conference on Electronics, Computer, and Computation (ICECCO), IEEE.
29 September 2018. pp. 1–4.
[6] Department of Public Health, Environmental and Social Determinants of

Health. UV radiation, World Health Organization. https://www.who.int/uv/
faq/whatisuv/en/index2.html, 9 March 2016.
[7] Buonanno M, Welch D, Shuryak I, and Brenner DJ. ‘Far-UVC light
(222 nm) efficiently and safely inactivates airborne human coronaviruses’.
Scientific Reports. 2000;10(1):1–8.
[8] Kitagawa H, Nomura T, Nazmul T et al. ‘Effectiveness of 222-nm ultra-
violet light on disinfecting SARS-CoV-2 surface contamination’. American
Journal of Infection Control. 2001;49(3):299–301.
[9] ProLampSales. Sailon UVC – 222 nm Far UV Light – 150 W – 455 mm,
ProLampSales. https://www.prolampsales.com/products/sailon-uvc-222nm-
far-uv-light-150w-455mm, 2021.
[10] Bhuyan HK, Chakraborty C, Shelke Y, and Pani SK. ‘COVID-19 diagnosis
system by deep learning approaches’. Expert Systems. 2021:e12776. https://
doi.org/10.1111/exsy.12776.
[11] Dash D, Chakraborty C, Giri SK, Pani SK, and Frnda J. ‘BIFM: Big-data
driven intelligent forecasting model for COVID-19’. IEEE Access.
2021;9:97505–17. 10.1109/ACCESS.2021.3094658.
[12] Aslam B, Javed AR, Chakraborty C, Nebhen J, Raqib S, and Rizwan M.
Computing. 2021:1–17. 10.1007/s00779-021-01596-3.
[13] Petrovic N and Kocic Ð. ‘IoT-based System for COVID-19 Indoor Safety
Monitoring (preprint)’. International Conference on Electrical, Electronic,
and Computing Engineering (IcETRAN). 2020. pp. 1–6.
[14] Karmore S, Bodhe R, Al-Turjman F, Kumar RL, and Pillai S. ‘IoT based
humanoid software for identification and diagnosis of Covid-19 suspects’.
IEEE Sensors Journal. 2020.
[15] Wanga H, Joseph T, and Chuma MB. ‘War against coronavirus (COVID–19)
in Tanzania: Designing a low cost automatic water tap’. International
Journal of Computer Applications Technology and Research. 2020;9
(04):150–4.
[16] Lastovicka-Medin G. ‘Social engineering and prototype awareness enhan-
cing during international coronavirus outbreak’. 9th Mediterranean
Conference on Embedded Computing (MECO), IEEE. 8 June 2020. pp. 1–4.
[17] Vasantharao G and Arifunneesa SK. ‘Temperature detection and automatic
sanitization and disinfection tunnel-COVID 19’. The International Journal
of Analytical and Experimental Modal Analysis. 2020;12:0886–936.
[18] Garg L, Chukwu E, Nasser N, Chakraborty C, and Garg G. ‘Anonymity
model’. IEEE Access. 2020;8:159402–14.
[19] Roy S and Chakraborty C. ‘Panic buying situation during COVID-19 global
pandemic’. Journal of Information Technology Management. 2021;
13(2):231–44.
[20] Dogan O, Tiwari S, Jabbar MA, and Guggari S. ‘A systematic review on AI/
ML approaches against COVID-19 outbreak’. Complex and Intelligent
Systems. 2021:1–24.
[21] Chakraborty C, Gupta B, and Ghosh SK. ‘A review on telemedicine-based
WBAN framework for patient monitoring’. Telemedicine and e-Health.
2013;19(8):619–26.
[22] Muhammad LJ, Algehyne EA, Usman SS, Ahmad A, Chakraborty C, and
Mohammed IA. ‘Supervised machine learning models for prediction of
2021;2(1):1–3.
[23] Ocaña J, Flores E, Naquin E, Pacheco A, and Felix E. ‘Smart system model
to disinfect SARS-CoV-2 in physical environments’. Journal of
Contemporary Issues in Business and Government. 2021;27(2).
[24] Megantoro P, Setiadi H, and Aji Pramudita B. ‘All-terrain mobile robot
disinfectant sprayer to decrease the spread of covid-19 in open area’.
International Journal of Electrical and Computer Engineering (IJECE).
2021;11(3):2090–100.
[25] Metcalfe B, Iravani P, Graham-Harper-Cater J, Bowman R, Stirling J, and
Wilson P. ‘A cost-effective pulse oximeter designed in response to the
COVID-19 pandemic’. Journal of Open Hardware. 2021;5(1).
[26] Reji B, Rathesh A, Suresh S, Jose N, and Benny J. ‘Smart Cap for Prevention
of Contagious Diseases and Social Distancing Using Arduino’. Bangalore
Humanitarian Technology Conference (B-HTC), IEEE. 8 October 2020.
pp. 1–5.
[27] Roy K. ‘Flexible (Portable) COVID 19 detecting device using Arduino and
sensors’. International Journal of Engineering Research in Computer
Science and Engineering. 2021;8(3):32–3.
[28] Raghav S, Vijay G, Harika PS et al. ‘Suraksha: Low Cost Device to
Maintain Social Distancing during CoVID-19’. 4th International Conference
on Electronics, Communication and Aerospace Technology (ICECA), IEEE.
5 November 2020. pp. 1476–80.
[29] Dhaya R. ‘Deep net model for detection of covid-19 using radiographs based
on roc analysis’. Journal of Innovative Image Processing (JIIP). 2020;
2(03):135–40.
[30] Singh S and Garg A. ‘Automatic door handle or knob sanitizer’. Journal on
Today’s Ideas-Tomorrow’s Technologies. 2020;8(1):35–40.
[31] Thakare MS and Kokate M. ‘A review on smart health system to monitor
people in Covid-19 quarantine’. International Journal of Advance Scientific
Research and Technology. 2020;5:28–33.
[32] Bhagawan GP and Purohit VV. ‘Safety regulations checks at entrance for
COVID safety’. International Journal of Research Publication and Reviews.
2021;2(3):383–6.
[33] Sharma K. ‘The ideas and innovations IITs have come up with to help India
fight coronavirus’. The print. 1 April 2020.
[34] Peters D, Esswein P, Krugel L, Obrycki V, Marcoux D, and Linderman T.

‘Far-UVC Applications in Healthcare’. University of Wisconsin-Madison,
Department of Biomedical Engineering BME. 7 October 2020. p. 27.
[35] Haider I, Ali A, Arifeen T, and Hassan AS. ‘Far UV-C lights and fiber optics
induced and selective far UV-C treatment against COVID-19 for fatality-
survival tradeoff’. TechRxiv, Powered by IEEE. 2017, Volume XX. pp. 1–9.
[36] Cole CA, FOrbes PD, and Davies RE. ‘An action spectrum for UV photo-
carcinogenesis’. Photochemistry and Photobiology. 1986;43(3):275–84.
[37] Lindblad M, Tano E, Lindahl C, and Huss F. ‘Ultraviolet-C decontamination
of a hospital room: Amount of UV light needed’. Burns. 2020;46(4):842–9.
[38] Seuylemezian A, Buonanno M, Guan L, Brenner DJ, and Welch D. ‘Far-
UVC light as a new tool to reduce microbial burden during spacecraft
assembly’. Advances in Space Research. 2001;67(1):496–503.
[39] Santosh Patnaik. UV-based coronavirus cleaner for gadgets arrives. The
Hindu; July 19 2020. https://www.thehindu.com/news/national/andhra-pra-
desh/uv-based-coronavirus-cleaner-for-gadgets-arrives/article31863658.ece,
10 May 2021.
[40] Raj A. UV Blaster, DRDO new UV Disinfection Equipment[ONLINE].
Infotonline; May 4 2020. https://infotonline.com/uv-blaster-drdo-new-uv-
disinfection-equipment/, 10 May 2021.
[41] Dimitrov A and Minchev D. ‘Ultrasonic sensor explorer’. 19th International
Symposium on Electrical Apparatus and Technologies (SIELA), IEEE. 29
May 2016. pp. 1–5.
[42] Prasojo I, Nguyen PT, and Shahu N. ‘Design of ultrasonic sensor and
ultraviolet sensor implemented on a fire fighter robot using AT89S52’.
Journal of Robotics and Control. 2020;1(2):59–63.
[43] Buonanno M, Welch D, Shuryak I, and Brenner DJ. ‘Far-UVC light
(222 nm) efficiently and safely inactivates airborne human coronaviruses’.
Scientific Reports. 2020;10(1):1–8.
[44] Nayyar A and Puri V. ‘A review of Arduino board’s, Lilypad’s & Arduino
shields’. 3rd International Conference on Computing for Sustainable Global
Development (INDIACom), IEEE. 16 March 2016. pp. 1485–92.
Chapter 14
Effect of COVID-19 pandemic on waste
management system and infection control
Ramkrishna Mondal1 and Chinmay Chakrabarty2
Abstract
Late within the year of 2019, the entire world roused by a reality of an epidemic of
coronavirus disease (COVID-19). It has become very vital and important to take all
possible precaution till the world finds a definite treatment against this disease as
there is least information about its behavior and origin. World Health Organization
(WHO) informed that proper guidelines need to be followed for COVID-19 waste
management as it is extremely infectious and contaminated. Personal protective
equipment has become a very vital element to guard from exposure of any infec-
tious materials. It is usually used in healthcare or hospital settings during any
outbreak. These enormous challenges are with all stakeholders for avoiding spread
through waste. Additionally, to WHO, each country was adopted different safety
measures and developed guidelines to control the contamination and manage the
waste. The previously mentioned guidelines are very useful for managing the
infectious waste and providing protection and security of waste handlers. Not only
the spread of COVID-19 is reduced by practicing appropriate technologies to
handle wastes, but it generates worth through increasing the recyclability chance of
waste. Due to COVID-19, there were lots of social and economic changes hap-
pened in the world, which leads to reduction in manufacturing units and various
business processes. All these have significantly affected waste management that is
a very vivacious step for the health outcomes of the patients as well as caregivers,
especially during the COVID-19 pandemic. Among all the issues created by
COVID-19, the problem that will create major mess is, not handling biomedical
waste properly. If the spreading needs to be controlled, it requires a strict mon-
itoring of the complete cycle starting from the point of generation. However, due to
the invaluable service of waste management sector only, it is possible to avoid
spread of the COVID-19 and ensure that the weird tons of waste will not be gath-
ered that increases health risks and the spread of disease.
1
Department of Hospital Administration, All India Institute of Medical Sciences Bhubaneswar,
Bhubaneswar, India
2
Department of ECE, BIT Mesra, Ranchi, India
Keywords: Biomedical waste; Coronavirus disease (COVID-19); Bio-

medical waste; Municipal solid waste; Waste management; Rethinking;
Effective waste management; Personal protective equipment
14.1 Introduction
Improper handling of waste, be it medical waste or healthcare waste or biomedical

waste (BMW) or household waste or municipal solid waste (MSW), will lead to
severe public health and environmental effects. Proper management of the said
wastes is mandatory to avoid any effect in environment and health protection.
Coronavirus disease (COVID-19) generates further challenges in waste manage-
ment, and significance of proper waste management should receive more attention
during this pandemic. COVID-19 spread rapidly all over China and it exposed the
deficits in BMW management. Subsequently, COVID-19 was declared an
International Public Health Emergency on January 31, 2020 by WHO as it rapidly
invaded due to high infection rate first in China, followed by other countries of the
world [1]. In the Wuhan city, the daily BMW generation ascended by five times,
250 t, from 40 t far exceeding its daily processing capacity of the BMW treatment
plant. Till first 2 months of 2020, a major number of new cases and death were
from China, whereas adjoining countries like Asia, North America [2], and Europe,
remained relatively safe [3] but later on all suffered a lot and now even some of
them are in the second wave phase. Both China and the United States during
COVID-19 had come up with a similar type of issues regarding BMW management
that were timely management of the waste. According to a WHO report, there is
hardly any evidence of transmission of COVID while handling BMW by the waste
handler who does not protect themselves properly [4]. The Solid Waste Association
of North America reported that if the sanitation workers maintain social distances
from other people, it will reduce their risk of getting infected with the virus [5]. The
need for personal protective equipment (PPE) was hastening worldwide for the
protection of doctors, nurses, healthcare workers, and waste handling staff, which
leads to a huge production of PPE. Subsequently, waste also increasing, which
could be either infected or not infected, due to lockdowns and the introduction of
home quarantines to control COVID-19 [6]. Although the pandemic had negative
impact on universal economic activities, it has made a lot of positive environmental
changes also [7]. The pandemic has made a significant impact not only on BMW
generation but also on MSW production and composition. As per the WHO and the
US Centers for Disease Control and Prevention (CDC), the waste generated during
pandemic may be classified into two categories: (i) within healthcare facilities and
(ii) outside healthcare facilities [8]. It can be said that COVID-19 has transformed
the waste production dynamic [9] with probable criticality in every step of the
waste management starting from the segregation, collection, transport, treatment,
and disposal [10]. In the pandemic, maintaining social distancing and other mea-
sures to protect from COVID-19 leads to increase in plastic use, a situation that
Effect of COVID-19 pandemic on waste management system 387
needs policy consequences [11]. This rising plastic use leads to a channel for
contamination and increases the chance of spread. To control the pandemic a pro-
hibition of movements was implemented, which affects the countries that depend
on external technologies for waste reprocessing and management. As a result, the
waste generated in that said period is disposed of rather than recycled. COVID-19
pandemic like emergencies made the sustainable management of BMW proble-
matic and amplified. The hefty waste either needs to be collected for disposing or
recycled. Both the processes are compromised because of the shortage of man-
power and hard work to implement infection control measures [12,13]. These dis-
rupted services lead to mishandling of waste and it increased up to 300% at few
rural communities of the United Kingdom (UK) [14]. Due to this pandemic the UK
Environment Agency allowed a temporary storage of waste and incineration ash at
sites that further threatens the environment [15]. The waste management system as
a whole needs to be considered, from waste generation, to treatment, including
resource use, and disposal of extra waste to reduce the socioeconomic and envir-
onmental impacts of waste as well as shielding of waste management chains will
help to achieve sustainable cities and communities [16]. In the last two decades,
COVID-19 pandemic is the third major zoonotic disease episode. First, it was the
SARS (severe acute respiratory syndrome) in 2002–03 and the second was MERS
(Middle East respiratory syndrome) in 2012. The disease was first reported to
WHO by the Chinese Health Officials on December 31, 2019 as atypical pneu-
monia of unknown reason [17]. According to Van Doremalen et al., the survival
period varies from hours to days depending on the materials and environment.
Survival phases on copper and cardboard vary from 3 to 4 h, while on plastic,
stainless steel varies from 1 to 3 days and some time as high as up to 9 days, even in
tap or waste water at 20 C for 2 days [18,19]. This longer survival period can lead
to a high danger of communal spread by inappropriate. In this chapter, first various
impacts like socioeconomic, environmental and also impacts on waste generation
and waste management are discussed. Different subsequent challenges faced due to
these impacts on waste management are enumerated. Lastly, it described the new
models or rethinking in the processes of waste management during pandemic.
14.2 Socioeconomic and environmental impact

Socioeconomically and demographically it indicates that COVID-19 is distressing
every person in a different way [20]. Due to the lack of suitable data, it is very
difficult to fully opine that why this COVID-19 affected different groups differ-
ently in the society, although poor have been affected most [21]. Moreover, local
and global business got disturbed severely by this pandemic [22]. It has been found
that even after taking corrective and preventive steps, a very less number of inter-
national communications started in China [23]. Different economic institutes and
World Bank are very much worried about the pandemic impact on economy. Loss
of job and business and go-slow of national transformations must be restored to
diminish the socioeconomic impact. After the pandemic started in, stock exchange
markets detected the major decline after 2008 economic disaster. The market slide
continues as global markets also fell down quite a lot. However, later slowly it
gained up after October/November 2020.
As a global pandemic, COVID-19 contributed a lot of environmental issues,
especially in BMW management. Wuhan city generates 200 t of BMW, four times
greater than the capacity of its waste treatment capabilities [24]. Ocean Asia, a
leading NGO, mentioned that BMW generated in Hong Kong spoiled the seaside
atmosphere [25]. Due to COVID-19, countries must announce lockdown measures
to safeguard their citizens and limit the exposure from COVID-19 [26]. Several air
changes were found around the world like air pollutants that came down in New
York by 50%, 25% reduction of emission in China due to 40% decrease of coal use
and a significant reduction of air pollution during pandemic [27]. Nevertheless,
financial influences of COVID-19 were very vital, but they highly contributed
toward an ecofriendly decarbonized environment, and air quality improved sig-
nificantly as reported in the various literature [28,29]. Satellite pictures of NASA in
order to analyses air quality show 20%–38% reduced pollutants in Korea, Spain,
Germany, and the USA [30].
14.3 Impact of waste generation
With the increasing number of patients, BMW generated from the COVID-19
patient wards are also increasing and becoming a bigger concern for hospital. Thus,
we have to ensure that this contaminated trash from hospitals should not cause any
harm to the public before they are safely disposed. Changes in BMW character-
istics are as follows: BMW includes the waste generated from hospitals, medical
institutions, medical laboratories, etc. where there are many patients who are
managed [31]. Generally, 75%–90% waste generated from hospital is general,
nonhazardous waste and does not require any special management, and only 10%–
25% are BMW and require special treatment for disposal. But, in a pandemic
situation, all the wastes generated within a facility would be measured as BMW and
managed with full precautions. To manage these unusual amount of BMW, the
existing waste treatment facilities need to be modified based on the real informa-
tion of the amount produced and availability of the treatment facilities [32]. A
report from UNEP says that hospitals generating waste 0.5 kg/bed/day before
pandemic increased the same during pandemic by 3.4 kg/bed/day. In India, waste
generation increased from 0.5 to 2.5–4 kg/bed/day as reported by Ramteke and
Sahu [33]. Also, in a study at a Chinese hospital by Peng et al. [10], it is reported
that the average waste generation rate is more than 6 kg/bed/day during pandemic.
Asymptomatic infections are caused by the COVID-19 virus [34,35], so that there
are obstacles in determining the level of COVID-19 transmission in communities
and between countries. A method that is quite effective in detecting and analyzing
the spread of virus is COVID-19 surveillance in wastewater as it provides unbiased
results and it can be used as an early warning system. Wastewater monitoring is a
virus identification system that has contaminated the environment and society;
reducing COVID-19 in wastewater can also be used as a marker of reducing

COVID-19 cases in the community or region [36]. A most recent study conducted
in the United States adopted a wastewater control approach. It was used to reveal
the phylogeny of the COVID-19 strain and to enable to evaluate the efficacy of
public health interventions to overcome outbreaks [37]. Waste generations in some
countries and regions were decreased due to many factors like lockdowns and
travel restrictions, especially in tourist cities like Barcelona by 25%; Macao, China,
by 17%–25%; and Catalonia, Spain by 17%. On the contrary, COVID-19 leads to
increase waste generation in some areas, like England, more than 90% higher than
usual [38]. The social-distancing measures reduced the generation of BMW but
shifted part of the BMW toward MSW [39]. Most of the developing countries were
facing the issues of poor management of BMW much before pandemic [40]. As per
the Asian Development Bank report, Manila, Kuala Lumpur, Hanoi, and Bangkok
produced 154–280 t of BMW daily during pandemic [41]. Detailed data of MSW
and BMW generation in different countries are shown in Table 14.1.
PPE Waste: Till today PPE is the most dependable and reasonably priced item
for defending contamination and spread of any viruses. In starting of the pandemic,
Table 14.1 Waste generation in different regions of the world
Decrease in waste generation Increase in waste generation

Country Waste generation Country Waste generation
Macao, China MSW decreased by England [38] MSW increases to
Source: Environmental 17%–25% by 0%–20%
Protection Bureau
Macao SAR
Morocco MSW generation Tehran, Iran [42] MSW increased by
Source: Ouhsine O decreased 34.7% and BMW
2%–10% increased by
18%–62%
Trento, Italy [9] MSW decreased by South Korea BMW increased by
14% (4,058 t/day) Source: MoE 2,600 t/day
Korea
Milan, Italy Total BMW and
Source: AMSA MSW decreased
by 27.5%
Catalonia, Spain MSW generation decreased by 17% (242,000 t/day)
Source: Generalitat de Catalunya but BMW increased by 350% (1,200 t/day)
and ACR
India [8] MSW decreased by about 20%–40% but BMW
Ahmedabad, India increased by 67%–82%, (1,000 kg/day)
Source: TOI
New York, USA MSW increased to 30% but commercial waste
Source: Waste Advantage decreased by 50%
Tokyo, Japan Commercial waste reduced by 57% but MSW
Source: UNEP increased by 110%
Table 14.2 Masks used by some countries
Country Mask used data

The Republic of Ireland 9 million masks/week
Source: Farsaci
Africa 700 million masks/day
Source: Nzediegwu and Chang
Iran [45] 5.5 million masks/day
Italy [9] 40 million masks/day
PPE was used chiefly by healthcare experts only, but as pandemic continues, PPE
has been used by general population very widely, which leads to increase the
demand of PPE during the pandemic. Almost every country has instructed man-
datory facemask use for its citizens. The mask uses of some countries are shown in
Table 14.2. A mask or PPE is largely composed of plastic (>80%) [43]. The
increase in demand and the mandatory masks use transformed the plastic waste
generation dynamics and enhanced intricacies of management of plastic and hin-
dered the hard work to decrease plastic trash [44].
Plastic waste: with the change in lifestyle, the use of plastics also increases for
convenience and reassurance. But at the same time concern about the reduction of
plastic waste was prevalent among everybody based on the world pollution due to
plastic. During this pandemic, there is a shift from concern about environmental
pollution of plastic to preferring plastic packaging [46]. Demand has risen abruptly
for packaging and transportation during this pandemic [47]. In most countries, to
control the spread quarantine caused an increase demand for online shop services
[44]. Overall, when compared to pre-pandemic, the online trading has increased in
pandemic and post pandemic era (16.6%) [48].
Food waste: COVID-19 has made changes in the life style and in the con-
sumption habits, which stimulate food loss and waste (FLW). In April 2020, a
survey of US households found, during pandemic, cooking at home increased by
54% (n ¼ 1,005) [49], which implies more food waste, whereas 57% conveyed less
waste, which means net effect with no change. Individuals waste more food, with
an anticipation of shortage of food during crisis, which results in excess food sto-
rage [50]. In Spain, consumption data shows increase in food purchases (29.8% and
10.9%) in first and second weeks and then decreased by 20.3% in the third week of
lockdown, which indicates that pandemic had very less effect on the FLW by 12%
hike that might be due to the socioeconomic crisis [51].
14.4 Impacts on waste management
As per the CDC, hazards of transmission of COVID-19 are high in persons who are
in close contact with an already diseased person or by touching a surface or object,
which is a less known mode of virus spreads. However, OSHA states that
acquaintance chance will be increased for the workers who deal with waste and
wastewater management. A WHO report showed that in 2000, 21 million people
were infected with hepatitis B, 2 million with hepatitis C, and 260,000 with HIV
because of polluted needle [52]. However, the waste thrown by isolated families
could lead the waste handler workers as well as cloth pickers to risk. It is further-
more crucial to ensure that waste should not be mixed at regular dumping grounds.
A good waste management stratagem denotes the program of a facility for mana-
ging produced waste for treatment and disposal. In response to the pandemic, every
country has implemented the best suitable management strategies based on their
resources [42,53]. There should be appropriate facilities to keep waste in temporary
storage in sealed container and in very secure areas, in the case of problem in
treatment and disposal methods. Disinfectant must use to avoid probable virus
transmission. Mobile incineration and autoclaves can be used to dispose the extra
BMW generated in pandemic. Designated vehicles should be used for transferring
waste with proper record [41] and special permit to collect healthcare waste for
smooth handling. The authorities should strictly oversee the process that regula-
tions are followed properly based on national and local guidelines. Any good
management system requires an assessment available options, plans, and clearly
defined roles and responsibilities of personnel. There should be a waste manage-
ment committee that will implement the plan. Strategies should be periodically
reviewed, and all staff involved in waste management processes should be updated
with the periodic changes [54]. Overall, the optimized use of resources and proper
waste segregation minimize the waste significantly [45]. As waste generation
cannot be predicted with increasing infections, temporary waste treatment tech-
nologies like autoclaves burn incinerators, it is difficult to manage waste effectively
[55]. Suitable waste management always helps in raising the quantity of recyclable
waste by the use of these alternative technologies during the pandemic [42].
Considering all these options, healthcare waste can be utilized to augment value to
the country’s economy during the pandemic [11,56].
Various international organizations published many policies or guidelines
related to waste management from time to time. Some countries have laws related
to BMW and some instantly issued new strategies and plans to overcome the
pandemic. Unexpected rise in BMW is a huge challenge for all. India also released
policies in time to time on BMW management [44]. In Iran, pandemic causes the
clearance of infectious waste to entirely stop for a certain period [45]. The National
Institute of Public Health, Romania, includes quarantine home generated waste as
infectious waste and implements firm waste management actions [57]. Diverse
decontamination and discarding methods and technologies are used based on fac-
tors like, quantity, type, and costs, some of them are given in Table 14.3 [29,56]. In
all-purpose, an amalgamation of chemical disinfection and incineration, the most
widely accepted, safe, simple and effective technology would be adopted [29]. In
the case of overloaded BMW treatment facilities, industrial furnaces or cement
kilns may be used safely [41]. On-site and mobile treatments are preferred due to
their flexibility in retorting to fluctuating demands. Fixed, mobile, coprocessing
facilities, and nonlocal disposal methods were all successively used in China,
Table 14.3 Disinfection and disposal techniques for biomedical waste (BMW)
Broad categories Specific technologies

Chemical Sodium/calcium hypochlorite, chlorine dioxide
Physical Microwave
High-temperature steam
Incineration Pyrolysis vaporization
Rotary kiln
Plasma
although all these are supplementary methods of techniques used in waste man-
agement [58].
14.4.1 Waste management adjustments

A survival period of viruses on diverse surfaces may vary going up to 9 days [28],
which represents that the waste from COVID patient could be a source of infection
and managing it will pose a risk to waste workers [9]. The different countries made
different adjustments in their waste management. Few common adjustments are
summarized next.
14.4.1.1 Healthcare waste/BMW

Whatever best practices available for safe healthcare waste management should be
followed, including enough human and material resources to segregate and dispose
the waste safely. All waste produced during patient care, including those with con-
firmed cases, is infectious and should be collected safely as per the national guideline
[59–61]. The waste moved out of hospital should be properly handled and treated
accordingly or disposed as per the law of the land. All healthcare waste handlers must
wear appropriate PPE and perform hand hygiene accordingly. Considering the out-
break, it is expected that infectious waste volume will increase and so of PPE waste,
therefore, to handle this additional waste treatment capacity that needs to be
increased. Alternative treatment technologies, like autoclaving or high temperature
incinerators, may need to be put in place to ensure their sustained operation. The ratio
of waste to infectious virus is much less in MSW compared to BMW, but the risk of
infection cannot be overlooked from any of the waste. Most of the countries agreed
that the recycling method carried very less risk and, thus, normal waste handling and
management can be processed like as before the pandemic. In case the waste is
polluted with infectious waste, it needs to be managed properly [62].
14.4.1.2 Household waste/MSW

There is no guideline available for household waste/MSW management by any
authorized body, but some basic measures should be followed. A waste bag may be
kept in the room for items used by the patient, and care-givers must remove their
gloves and masks immediately after use in another bag and should be kept near
door during leaving. Waste bags must be closed before being removed and never be
emptied in another bag. Strict hand hygiene should be maintained after handling
waste bag.
14.4.1.3 PPE waste

In the UK, for quarantine patients, PPE waste needs to be put into a black bag waste
bin and can be permitted to store for 72 h. The Portuguese Environment Agency
recommends that all PPE waste generated by ordinary citizens must be put in a
closed and watertight trash bags and disposed preferably by incineration or landfill
as opined by Patrı́cio Silva et al. [63]. China did not issue any guidelines related to
PPE waste. A survey by OCEANSASIA reported that a huge number of surgical
masks were found lying on Soko Islands.
14.4.1.4 Dead bodies

Although the danger of virus spread from dead bodies is very less, workers hand-
ling dead bodies must follow standard precautions. The mortuary staff preparing or
handling the body must wear proper PPE, including impermeable apron, and PPE
must be judiciously removed and disposed as per protocol, and hand hygiene must
be performed. The dead body of a confirmed or suspected case should be wrapped
properly and transferred to the mortuary area as early as possible.
14.5 Challenges in handling waste
Worldwide, the spread of COVID-19 has amplified the use of PPE manifold, and
all these littered PPE waste from healthcare and home might be a vector for COVID
virus, due to the potential ability of survival of the virus even up to 7 days on
masks, gloves, and fomite. Additionally, a polymer plastic, polypropylene, takes
long time to degrade and likely culminate in further plastic pollution as reported by
WHO [52]. A COVID-19 patient can generate 3.4-kg waste per day that shows the
gradual increasing trend in waste quantity in pandemic [41]. The BMW from the
healthcare institution and hospital in India is generating 517 t/day. Out of 517 t/day,
only 501 t/day waste could be treated in Common Biomedical Management Waste
Treatment Facility over 28 states of India. The rest 16 t/day wastes are accumulated
untreated daily. Some states like Goa, Andaman Nicobar, Mizoram, Nagaland, and
Sikkim are not having CBWTFs at all [44]. Many other countries are also facing
alike challenges to manage the vast quantity of waste [41]. Healthcare solid waste
in France increased from 40% to 50% and in the Netherlands, it increased to 50%
from 40% similar to India and Iran as reported by Zand and Heir [42] and Prata
et al. [44]. European administrations are facing challenges in retaining waste
management staff, providing them a good atmosphere and space for extra waste
generated in pandemic. In many of the developing countries, there are no effective
BMW management policies and resources. South Asia is a thickly populated sec-
tion of the world with 24.89% of the global population. Safe and accurate BMW
management is a legitimate necessity for such areas to avoid a further health crisis.
Also, the urban–rural inequality in waste management is a great challenge. The
INCLEN Program Evaluation Network Study Group reported that a big section of
healthcare system (82% of primary, 60% of secondary, and 54% of tertiary care) is
lacking in proper waste-handling system [4]. In rural areas, informal practitioners
with no or very little practical knowledge handle BMW, and awareness of BMW
segregation and management is more different among staff in urban than in rural
areas. Lots of studies recommended that these previously mentioned countries need
to formulate comprehensive guidelines to seal knowledge gaps in the BMW
handling. Developing countries have neglected BMW management in obedience
with rules for a long time. The discussed issue of indiscriminate dumping of waste
will pose a severe environmental impact and occupational health risks. Besides,
ignorant public conduct will put waste handlers at risk as the quarantined domestic
litters are disposed with the regular household MSW. During these pandemics, like
India, many countries have come up with some thought-provoking problems that
include the following:
● The significantly increasing quantity of hazardous elements led to tricky issues
of separate waste collection.
● During pandemic, no formal training was given to sanitary workers on how to
handle the waste. Arranging regular training to the operators is also a
challenging issue.
● Irregularity in supply of PPEs and disinfectants increase the probabilities of
collection staff who are prone to get infected.
● Inappropriate donning and doffing of the PPEs among waste worker may
increase spread between them.
● Not keeping the social-distancing measures at collection centers and
treatment areas.
● Wastewater of healthcare facilities may contain virus, staff who are handling
wastewater treatment plants may at a high risk of infection.
● No door-to-door collecting staff are there from residential areas, and the one-
point collection is predominant for increasing the infection risk.
● Use of separate vehicles for infected and noninfected waste transport and lack
of disinfectant to clean the vehicles increase the chance of spreading the virus.
● Lack of proper monitoring review and verification systems in towns and villages.
● Challenges in implementing strategies and advisory at the ground level people
who handling waste.
● Difficulty to make general public aware within short span of importance and
rules of segregating the waste.
● At few places, manual loading or unloading of the waste is required, which
raises the probabilities of contaminations.
14.6 Rethinking effective waste management

WHO forecasted that the pandemic will continue for an extended time and it will
have a longstanding effect on environmental sustainability. Accordingly, the whole
world is focusing on prevention plans and vaccine research. Most of the countries
were accustomed with or restructured their strategies of waste management for the
COVID-19 outbreak. In this connection, China can offer valued inputs considering
their experiences of building an all-inclusive disposal system through a mixture of
central disposal as well as on-site emergency disposal of waste (like mobile treat-
ment and industrial kilns). The pandemic not only reformed people’s lifestyle but
stirred extra contemplation on the human and nature relationship. In addition, there
are few critical issues about whether the existing systems of BMW management
can quickly restore the waste processing, and ability of general population to focus
on waste reduction. These issues need to be discussed and studied for betterment in
the future. There is not a solitary resolution available for different countries, which
varies in terms of infested cases, health resources; therefore diverse resolutions
should be implemented. There are lots of issues in dealing with appropriate and
actual waste management as it involves many departments and many individuals,
requiring interdepartmental collaboration. For a country to determine whether it
can manage a substantial rise in BMW production, it should analyze its existing
strategies, rules, structures, and also the present operation status of waste man-
agement. The main issue is not how to frame the strategies and policies, rather how
successfully can they be implemented. To creating a sustainable waste management
system throughout and subsequently the pandemic at national or local levels, both
strategy makers and experts need to consider at least seven thematic significance
areas as mentioned by UNICEF [4].
14.6.1 Policy, regulatory, and guidelines

Although International principles and strategies to managing BMW are available
and widely referred to and followed by countries, this only offers a decent foun-
dation in healthcare waste management during the COVID-19 pandemic.
Supplementary rules and guiding principle will always be necessary based on the
country-specific need and available resources to control the waste generated from
both healthcare and non-healthcare facilities, especially when there is an increased
generation of infected wastes. During an emergency or pandemic time, a robust
arrangement is required with clearly defined roles and responsibilities assigned for
waste management, including exigency plan, preparedness for further improvement
through review of existing policies. It is very important to conduct a quick valua-
tion of the existing facility systems of the country or city to detect the gaps and how
to increase the usage of available facilities toward maximum capacity.
14.6.2 Handling of infectious waste

It is vital to have knowledge about the huge healthcare waste production in the
pandemic, but at the same time, scientific understandings, like the viability of the
virus, should also be paid attention to make a sound waste management system [4].
It needs to be ensured for the appropriate segregation, packaging, and storage of
possibly contaminated wastes. The collection regularity should be based on priority
and the type of waste. A proper use of PPE and hand hygiene along with precau-
tionary practices to protect the health of waste employees should be maintained.
14.6.3 Suitable disposal methods

Selection of appropriate treatment choices is relative and required to study
numerous aspects. It is not limited to national and international principles only, but
also ecological and work-related safety, waste characteristics, suitable technologies
expenses, maintenance factors, etc. are very important. In reality, in developing
countries, there are many facilities that are lacking most desirable technologies for
waste treatment. In this situation, it is suggested that such countries should have at
least temporary or interim treatment methods like single-chamber incineration or
automated pressure pulsing gravity autoclaves, etc. In emergency pandemic,
developing countries can put on these interim treatment options, in harmony with
the national and international guiding principles. As a sustainable development
concept, coprocessing, open burning, and temporary storage as a momentary
measure may be considered [41].
14.6.4 Information, education, and communication

Healthcare workers are habitually skilled in handling healthcare waste, but capacity
building and awareness program is now very much necessary for households and
other nonmedical public. Program using a user-friendly communication channel
such as radio, television, and websites with everyday displays can intensify public
awareness on safe waste handling.
14.6.5 Data management and learning

For planning and policy development, the first step is a collection of data. Thus, the
importance of data as is immense. However, the availability of the data is often
very difficult, particularly during an emergency as resources are allocated during
emergency on urgent responses. Help of academic institutions to gather data is a
very useful option. Regular discussion is a good practice and is needed to exchange
views and lessons learnt for continuous improvement during any emergency
situation, specially in pandemic.
14.6.6 Monitoring of segregation

There should be a monitoring of the first and foremost step of the biomedical waste
management, i.e. segregation that limits the waste generation as well as the spread of
infection. Biomedical waste segregation deficiency index (BMWSD index) may be
used to monitor the segregation that could be used for training need and corrective and
preventive action as discussed by Mondal and Satyanarayana [64,65]. The study
shows how BMWSD index improves over 3 months, depicted in Figure 14.1.
14.6.7 Basic principles for managing waste

During pandemic, some basic ideologies and guidelines were followed to manage
infectious waste during infectious disease pandemic.
Waste generated through healthcare facilities (BMW) is as follows: these types
of waste should be achieved using national and international guidelines. Enhanced
Comparison of monthly over all entire Hospital Avg.

BMW Index.
BMWSD index 1.25
1.219
1.2 1.18
1.15 1.138 BMWSD index
1.1
1.05
June 2012 July 2012 August 2012
Months
Figure 14.1 Comparison of monthly improvement in average BMWSD index over

all entire hospital
treatment and disposal capacities in epidemic areas should be identified based on

the analysis of the gaps in capacity for waste, options available, mode of treatment
and disposal, local constraints, and considering infection concerns for human health
and environmental impacts.
Waste generated through households is as follows: priority should be given to the
steadiness of existing waste management services. Contingency planning includes
securing availability of necessary human resources and equipment. Special attention is
mandatory to alleviate dangers to people doing the waste operations. Reinforce
awareness amongst waste management service providers guarantees an adequate
supply of materials, and training, and capacity building of the staff.
14.6.8 Fund raising and national and international

collaboration
The local government should realize the need of fund-raising activities importance
in relation to the BMW management. In future, to tackle such pandemic situation,
sufficient fund should be maintained. National and international level collabora-
tions with various agencies are necessary for such abovementioned activities.
The author proposed a new model of waste management during pandemic. First,
there should be a detailed assessment or planning of the capacity of the both waste
generators and waste handlers. These will help use to plan properly. Out of the three Rs
(reduce, reuse, and recycle), first two could not be applicable during COVID-19 pan-
demic but with suitable technology, recycle must be tried. The main difficulty was the
shortage of incinerators. Suggestion for establishment of new incinerators should be
thought of to make big hospitals and be self-sufficient with a provision of extra capacity
if needed, although the space constrain should be considered and may be relaxed if
feasible. More staff training to handle waste keeping all possible precautions is required.
Mobile incinerator and industrial incinerators may also be used during pandemic.

COVID-19 made a high impression on economic development and health out-
comes. Apart from the economic slowdown, COVID-19 majorly affects the daily
wage and migrant workers [66,67]. It is now considered waste management as an

essential public service and it needs an urgent attention of proper waste manage-
ment from households, medical facilities. The contaminated waste may taint the
waste handlers as they are directly exposed to waste and use poor safety devices,
and this may lead to gradual increase of virus spread. Appropriate waste manage-
ment also helps in recycling of the waste or renovates the waste into a valuable
product like energy, which will complement worth to economies aimed at sus-
tainable development. The most important challenges with the management of
COVID-19 are its safe disposal from the point of generation to treatment facilities.
There is a great concern on the managing of COVID-19 waste, because the positive
cases increase rapidly, and the hospitals are overburdened by such BMW and if
precautions do not seem to be ensured then could end in contaminations among
sanitary workers. Proper and systematic sanitization practices and adequate PPE for
workers involved in handling and collection of BMWs should be mandatory. The
staff should be provided with proper and regular training to handle the waste and
stop transmission of virus through the waste. Proper segregation in healthcare waste
is mandatory for any healthcare organization, which should be monitored in a strict
manner. In these, connection matrix methods of monitoring to assess training need
assessment with the use of BMWSD index that may be adopted [64,65]. Public
awareness for waste segregation at the point of generation at homes before it can
harm the other masses should be promoted, which will help to segregate the waste
utilized in homes. As PPE is regularly used by everyone, which is made up of
plastic, forthcoming research should oriented towards biodegradable environment-
friendly protecting items. Regular monitoring is essential, while eliminating
COVID-19 waste with the aim that the workers should not get infected by it. The
communities at large, who are at home quarantine, must manage their waste
properly to make sure the protection of municipal workers and ragpickers. Proper
safety kits and training must be used for municipal workers engaged in handling
during the outbreak. There are lots of challenges, but suitable strategies and careful
handling can alleviate the hazards a lot.
References
[1] Bashir MF, Ma B, and Shahzad L. A brief review of socio-economic and
environmental impact of Covid-19. Air Quality, Atmosphere & Health 2020;
13:1403–1409. doi:10.1007/s11869-020-00894-8. Epub ahead of print.
PMID: 32837620; PMCID: PMC7395801.
[2] Bashir MF, Ma B, Bilal, et al. Correlation between climate indicators and
COVID-19 pandemic in New York, USA. Science of the Total Environment
2020; 728:138835. https://doi.org/10.1016/j.scitotenv.2020.138835.
[3] Chakraborty I and Maity P. COVID-19 outbreak: migration, effects on
society, global environment and prevention. Science of the Total
Environment 2020; 728:138882. https://doi.org/10.1016/j.scitotenv.2020.
138882.
[4] WHO and UNICEF. Water, sanitation, hygiene and waste management for
the COVID-19 Virus. Interim Guidance, WHO/2019-nCoV/IPC_WASH/
2020.2. [Online] April 23, 2020; 1–6. [Cited: April 15, 2021] https://apps.
who.int/iris/bitstream/handle/10665/331846/WHO2019-nCoV-IPC_WASH-
2020.3-eng.pdf.
[5] Calma J. The COVID-19 pandemic is generating tons of medical waste.
[Online] Mar 26, 2020. [Cited: April 15, 2021] https://www.theverge.com/
2020/3/26/21194647/the-covid-19-pandemic-is-generating-tons-of-medical-
waste.
[6] Lal P, Kumar A, Kumar S, et al. The dark cloud with a silver lining:
Assessing the impact of the SARS COVID-19 pandemic on the global
environment. Science of the Total Environment 2020; 732:139297.
[7] Espejo W, Celis JE, Chiang G, and Bahamonde P. Environment and
COVID-19: Pollutants, impacts, dissemination, management and recom-
mendations for facing future epidemic threats. Science of the Total
Environment 2020; 747:141314. doi:10.1016/j.scitotenv.
[8] Somani M, Srivastava AN, Gummadivalli SK, and Sharma A. Indirect
implications of COVID-19 towards sustainable environment: An investiga-
tion in Indian context. Bioresource Technology Reports 2020; 11:100491.
[9] Ragazzi M, Rada EC, and Schiavon M. Municipal solid waste management
during the SARS-COV-2 outbreak and lockdown ease: Lessons from Italy.
Science of the Total Environment 2020; 745:141159.
[10] Peng J, Wu X, Wang R, Li C, Zhang Q, and Wei D. Medical waste man-
agement practice during the 2019–2020 novel coronavirus pandemic:
Experience in a general hospital. American Journal of Infection Control
2020; 48(8):918–921.
[11] Klemeš JJ, Fan YV, Tan RR, and Jiang P. Minimising the present and future
plastic waste, energy and environmental footprints related to COVID-19.
Renewable and Sustainable Energy Reviews 2020; 127:109883. https://doi.
org/10.1016/j.rser.202.
[12] European Commission. Waste Management in the Context of the
Coronavirus Crisis. Brussels: European Commission, 2020.
[13] Association of Cities and Regions for Sustainable Resource Management.
Municipal waste management and COVID-19. Municipal waste manage-
ment and COVID-19. Association of Cities and Regions for Sustainable
Resource Management, 2020.
[14] Roberts K, Stringfellow AM, and Williams ID. Rubbish is piling up and
recycling has stalled – waste systems must adapt. The Conversation, 2020.
Available from: https://theconversation.com/rubbish-is-piling-up-and-recy-
cling-has-stalled-waste-systems-must-adapt-137100.
[15] UK Environment Agency. COVID-19 and storing waste at unpermitted sites
due to exceeding your storage limits: RPS C17. UK Environment Agency,
2020.
[16] United Nations. Sustainable development goal 11: Make cities and human
settlements inclusive, safe, resilient, and sustainable. United Nations, 2020.
[17] Poon LLM and Peiris M. Emergence of a novel human coronavirus threa-
tening human health. Nature Medicine 2020; 26(3):317–319.
[18] van Doremalen N, Bushmaker T, Morris DH, et al. Aerosol and surface
stability of SARS-CoV-2 as compared with SARS-CoV-1. The New England
Journal of Medicine 2020; 382(16):1564–1567.
[19] Wang D, Hu B, Hu C, et al. Clinical characteristics of 138 hospitalized
patients with 2019 novel coronavirus–infected pneumonia in Wuhan, China.
JAMA 2020; 323(11):1061–1069. https://doi.org/10.1001/jama.
[20] Collivignarelli MC, Abbà A, Bertanza G, Pedrazzani R, Ricciardi P, and
Miino MC. Lockdown for CoViD-2019 in Milan: what are the effects on air
quality? Science of the Total Environment 2020; 732:139280. https://doi.org/
10.1016/j.scitotenv.2020.139280.
[21] Messner W. The institutional and cultural context of cross-national variation
in COVID-19 outbreaks. medRxiv 2020. https://doi.org/10.1101/2020.03.30.
20047589.
[22] Goodell JW and Huynh TLD. Did congress trade ahead? Considering the
reaction of U.S. industries to COVID-19. Finance Research Letters 2020;
36:101578. https://doi.org/10.1016/j.frl.2020.101578.
[23] Bogoch II, Watts A, Thomas-Bachli A, Huber C, Kraemer MU, and Khan K.
Pneumonia of unknown etiology in Wuhan, China: Potential for interna-
tional spread via commercial air travel. Journal of Travel Medicine 2020; 27.
https://doi.org/10.1093/jtm/taaa008.
[24] Luan PT and Ching CTS. A reusable mask for coronavirus disease 2019
(COVID-19). Archives of Medical Research 2020; 51:455–457. https://doi.
org/10.1016/j.arcmed.2020.04.001.
[25] Hellewell J, Abbott S, Gimma A, et al. Feasibility of controlling COVID-19
outbreaks by isolation of cases and contacts. The Lancet Global Health
2020; 8:e488–e496. https://doi.org/10.1016/S221.
[26] Zambrano-Monserrate MA and Ruano MA. Has air quality improved in
Ecuador during the COVID-19 pandemic? A parametric analysis. Air
Quality, Atmosphere & Health 2020; 13:929–938. https://doi.org/10.1007/
s11869-020-00866-y.
[27] Gautam S. COVID-19: air pollution remains low as people stay at home. Air
Quality, Atmosphere & Health 2020; 13:853–857.https://doi.org/10.1007/
s11869-020-00842-6.
[28] Conticini E, Frediani B, and Caro D. Can atmospheric pollution be con-
sidered a co-factor in extremely high level of SARS-CoV-2 lethality in
northern Italy? Environmental Pollution 2020; 261:114465. https://doi.org/
10.1016/j.envpol.2020.114465.
[29] Sharif A, Aloui C, and Yarovaya L. COVID-19 pandemic, oil prices, stock
market, geopolitical risk and policy uncertainty nexus in the U.S. economy:
Fresh evidence from the wavelet-based approach. The International Review of
Financial Analysis 2020; 70:101496. https://doi.org/10.1016/j.irfa.2020.101496.
[30] Knowland EK, Keller C, Ott L, et al. Local to global air quality simulations
using the NASA GEOS composition forecast model: GEOS-CF, 2020.
[31] Rupani PF, Nilashi M, Abumalloh RA, Asadi S, Samad S, and Wang S.
Coronavirus pandemic (COVID-19) and its natural environmental impacts.
International Journal of Environmental Science and Technology 2020;
17:4655–466.
[32] Sarkodie SA and Owusu PA. Global assessment of environment, health and
economic impact of the novel coronavirus (COVID-19). Environment,
Development and Sustainability 2020: 1–11.
[33] Ramteke S and Sahu BL. Novel coronavirus disease 2019 (COVID-19)
pandemic: Considerations for the biomedical waste sector in India. Case
Studies in Chemical and Environmental Engineering 2020; 2:100029. doi.
org/10.1016/j.cscee.2020.100029.
[34] Nishiura H, Kobayashi T, Suzuki A, et al. Estimation of the asymptomatic
ratio of novel coronavirus infections (Covid-19). International Journal of
Infectious Diseases 2020; 94:154–155. https://doi.org/10.1016/j.ijid.20.
[35] Mizumoto K, Kagaya K, Zarebski A, and Chowell G. Estimating the
asymptomatic proportion of coronavirus disease 2019 (Covid19) cases on
board the diamond princess cruise ship. Eurosurveillance 2020; 25(10):1–5.
https://doi.org/10.2807/15607917.es.2020.25.10.20.
[36] O’Bannon D., Xagoraraki I, and O’Brien E. Wastewater-based
Epidemiology for Early Detection of Viral Outbreaks. Women in Water
Quality. Springer Nature Switzerland, 2020; 75–79. https://doi.org/10.1007/
978-3-030-17819-2.
[37] Nemudryi A, Nemudraia A, Surya K, et al. Temporal detection and phylo-
genetic assessment of SARS-CoV-2 in municipal wastewater. Cell Reports
Medicine 2020; 1(6):10098. https://doi.org/10.1016/j.xcrm.2020.100098.
[38] ADEPT. COVID 19 – Waste survey results, week commencing 18th May.
[Online] ADEPT, 2020. www.adeptnet.org.uk/documents/covid-19-waste-
survey-results-wc-18th-may.
[39] Naughton C. Will the COVID-19 pandemic change waste generation and
composition? The need for more real-time waste management data and systems
thinking. Resources, Conservation and Recycling 2020; 162:105050.
[40] Rahman MM, Bodrud-Doza M, Griffiths MD, and Mamun MA. Biomedical
waste amid Covid19: Perspectives from Bangladesh. The Lancet Global
Health 2020; 8(10):e1262. https://www.thelancet.com/journals/langlo/arti-
cle/PIIS2214-109X(20)30349-1/abstract.
[41] Asian Development Bank (ADB). Managing infectious medical waste dur-
ing the COVID-19 pandemic. [Online] Asian Development Bank (ADB),
2020. www.adb.org/publications/managing-medical-wastecovid1.
[42] Zand AD and Heir AV. Emerging challenges in urban waste management in
Tehran, Iran during the COVID-19 pandemic. Resources, Conservation and
Recycling 2020; 162:105051.
[43] Prata JC, Silva ALP, Walker TR, Duarte AC, and Rocha-Santos T. COVID-
19 pandemic repercussions on the use and management of plastics.
Environmental Science & Technology 2020; 54(13):7760–7765.
[44] Vanapalli KR, Sharma HB, Ranjan VP, et al. Challenges and strategies for
effective plastic waste management during and post COVID-19 pandemic.
Science of the Total Environment 2021; 750:141514.
[45] Rodrı́guez-Pardo J, Fuentes B, Alonso de Leciñana M, et al. Acute stroke
care during the COVID-19 pandemic. Ictus Madrid Program recommenda-
tions. Neurologia 2020; 35:258–263.
[46] Grodzińska-Jurczak M, Krawczyk A, Jurczak A, Strzelecka M, Rechciński
M, and Boćkowski M. Environmental choices vs. Covid-19 pandemic fear –
Plastic governance re-assessment. Socialist Register 2020; 4(2):49–66.
[47] Feber D, Lingqvist O, and Nordigården D. How the packaging industry can
navigate the coronavirus pandemic. mckinsey.com. [Online] 2020. www.
mckinsey.com/industries/paper-forest-productsand-packaging/our-insights/
how-the-packaging-industry-can-n.
[48] Clement J. Coronavirus global online traffic impact as of September 2020.
[Online] September 2020. www.statista.com/statistics/1105495/coronavirus-
traffic-impact.
[49] Hunter. Food Study Special Report. [Online] slideshare.net, 2020. www.
slideshare.net/HUNTERNY/hunter-food-study-special-report-america-gets-
cooking-231713331.
[50] Brizi A and Biraglia A. “Do I have enough food?” How need for cognitive
closure and gender impact stockpiling and food waste during the COVID-19
pandemic: A cross-national study in India and the United States of America.
Personality and Individual Differences 2021; 168:110396.
[51] Jribi S, Ben Ismail H, Doggui D, and Debbabi H. COVID-19 virus outbreak
lockdown: What impacts on household food wastage?. Environment,
Development and Sustainability 2020; 22(5):3939–3955.
[52] World Health Organization. Safe Healthcare Waste Management. Policy
Paper. Department of Protection of the Human Environment Water,
Sanitation and Health. Geneva: World Health Organization, 2016.
[53] Singh N, Tang Y, Zhang Z, and Zheng C. Covid19 waste management:
Effective and successful measures in Wuhan, China. Resources,
Conservation & Recycling 2020; 163(105071):1–3. https://doi.org/10.1016/j.
resconrec.2020.105071.
[54] Thakur V. Framework for PESTEL dimensions of sustainable healthcare
waste management: Learnings from COVID-19 outbreak. Journal of
Cleaner Production 2021; 287:125562.
[55] Yu H, Sun X, Solvang WD, and Zhao X. Reverse logistics network design
for effective management of medical waste in epidemic outbreaks: insights
from the coronavirus disease 2019 (COVID-19) outbreak in Wuhan (China).
International Journal of Environmental Research and Public Health 2020;
17:1770.
[56] Ilyas S, Srivastava RR, and Kim H. Disinfection technology and strategies
for COVID-19 hospital and bio-medical waste management. Science of the
Total Environment 2020; 749:141652.
[57] Mihai FC. Assessment of COVID-19 waste flows during the emergency state
in Romania and related public health and environmental concerns.
International Journal of Environmental Research and Public Health 2020;
17(15):5439.
[58] Yang L, Yu X, Wu X, et al. Emergency response to the explosive growth of
health care wastes during COVID-19 pandemic in Wuhan, China.
Resources, Conservation and Recycling 2021; 164:105074.
[59] Bhuyan HK, Chakraborty C, Shelke Y, and Pani SK. COVID-19 diagnosis
system by deep learning approaches. Expert Systems 2021:e12776. https://
doi.org/10.1111/exsy.12776.
[60] Dash S, Chakraborty C, Giri SK, Pani SK and Frnda J. BIFM: big-data
driven intelligent forecasting model for COVID-19. IEEE Access 2021: 1–
13. 10.1109/ACCESS.2021.3094658.
[61] Aslam B, Javed AR, Chakraborty C, Nebhen J, Raqib S, and Rizwan M.
Blockchain and ANFIS empowered IoMT application for privacy preserved
contact tracing in COVID-19 pandemic. Personal and Ubiquitous
Computing 2021; 1–17. 10.1007/s00779-021-01596-3.
[62] Ikiz E, Maclaren VW, Alfred E, and Sivanesan S. Impact of COVID-19 on
household waste flows, diversion and reuse: The case of multi-residential
buildings in Toronto, Canada. Resources, Conservation and Recycling 2021;
164:105111.
[63] Patrı́cio Silva AL, Prata JC, Walker TR, et al. Increased plastic pollution due
to COVID-19 pandemic: Challenges and recommendations. Chemical
Engineering Journal 2021; 405:126683.
[64] Mondal R and Satyanarayana P. Bio-Medical Waste Segregation Monitoring
in Healthcare Establishment – An Indian Scenario. LAP LAMBERT
Academic Publishing, 2018. ISBN-10: 6139941172; EAN:9786139941179.
[65] Mondal R and Satyanarayana P. A study on bio-medical waste segregation
monitoring in a tertiary care hospital at Andhra Pradesh. Asian Journal of
Management 2017; 8(1):73–77. doi:10.5958/2321-5763.2017.000117.
[66] Roy S and Chakraborty C. Panic buying situation during COVID-19 global
pandemic. Journal of Information Technology Management 2021; 13
(2):231–244. doi:10.22059/jitm.2021.80625.
[67] Chakraborty C, Roy S, Sharma S, and Tran TA. Environmental
Sustainability for Green Societies: The Impact of the COVID-19 Pandemic.
Springer Nature, 2020. ISBN: 978-3-030-66489-3.
Chapter 15
Natural adjunctive therapies options other than
COVID-19 antiviral therapies
Betul Ozdemir1 and Zeliha Selamoglu2
Abstract
The coronavirus disease 2019 (COVID-19) pandemic is the largest health problem
worldwide. Unfortunately, the lack of an effective and clear treatment causes it to
be a major health problem. There are currently no effective antiviral drugs or
vaccines. The symptoms and course of the disease differ individually. Symptoms
vary from asymptomatic to intensive care, even death. The individual variation of
this symptom pattern is related to viral load, individual’s current comorbid condi-
tions, age, gender, and most importantly, immune status. Considering the course
differences of the disease in all these individual, familial, and demographic dis-
tributions, it suggests that genetic and environmental factors play an important role.
There is a systemic inflammatory response in COVID-19. High levels of chemo-
kine and proinflammatory cytokines are detected in patients. Nutrition is one of the
most important factors for health. With the support of the immune system, people
can be protected from COVID-19 and make the process easy when suffering from
disease. Apart from the current treatments, some herbal and natural products are
used as adjunctive therapy. Key dietary components such as vitamins C, D, E, zinc,
selenium, nonflavonoids, flavonoids, polyphenols, and curcumin have been shown
to have immunomodulatory properties, which can help with infectious illnesses.
Most of these nutrients also have been demonstrated to be useful in the treatment of
COVID-19. The importance of such dietary elements in immunity and their parti-
cular consequences in COVID-19 patients are discussed in this review.
Keywords: COVID-19; Immunmodulation; Adjunctive therapies; Vitamin

D; Vitamin C; Vitamin E; Zinc; Selenium; Nonflavonoids; Flavonoids;
Polyphenols; Curcumin
1
Department of Cardiology, Faculty of Medicine, Niğde Ömer Halisdemir University, Niğde, Turkey
2
Department of Medical Biology, Faculty of Medicine, Niğde Ömer Halisdemir University, Niğde,
Turkey
15.1 Introduction
The 2019 novel coronavirus (2019-nCoV) or it is now referred to as the severe
acute respiratory syndrome coronavirus 2 (SARS-CoV-2) that origins in Wuhan
City of Hubei Province of China then is rapidly spreading the rest of the world and
causes the coronavirus disease 2019 (COVID-19) [1]. In this pandemic period,
which affects many individuals, some of the burdens on the health system should be
relieved with easy methods [2,3]. Many people with similar symptoms of fever,
cough, and shortness of breath applied to health institutions in Wuhan around
December 12, 2019 [4]. On March 11, 2020, it was announced to the whole world
that COVID-19 was a pandemic. In order to take precautions in pandemics, it is
necessary to know the origin of the pathogen, its hosts and evolution. The exact
origin of SARS-CoV-2 is still unclear. It has been determined that the genetic
structure of SARS-CoV-2 and the genetic structure of the bat SARS-like corona
virus are 96% similar. However, new coronavirus genomes have been shown in
pangolins showing 85.5%–92.4% similarity to SARS-CoV-2. In the animal king-
dom, only bats and pangolins have been found to be infected with SARS-CoV-2-
associated coronaviruses [5]. Coronaviruses have a structurally enveloped genetic
stalk containing a positive single-stranded RNA [6]. Coronaviruses are divided into
four subfamilies: alpha-, beta-, gamma-, and delta-coronaviruses. SARS-CoV-2 is a
member of the beta-coronavirus family. Its anatomy has been recognized as spike
(S) protein, membranes (M) protein, and envelope (E) protein. The S protein atta-
ches to the cell and allows the parasite to infiltrate the host. The N protein, which is
a member of the helix nucleocapsid holding genomic RNA, is just another element
of beta-coronavirus [7]. SARS-CoV-2 transmission is known to occur through
respiratory droplets and contact pathways. Transmission occurs through droplets
from all mucous membranes such as the mouth, nose, and eyes. Contact with sur-
faces around an infected person causes indirect transmission. Transmission may
also occur during the presymptomatic incubation period [8]. Follow-up of contacts
is as important for infection control as the diagnosis of infection [9]. After an
incubation period of approximately 5.2 days appear the symptoms of COVID-
19 infection. The period from the beginning with an average of 14 days is expected
for COVID-19 symptoms, and also the patient’s immune system response and
average of the age of the patient effect on that period are observed. It causes
contamination in asymptomatic patients [10]. It has been shown that symptoms
generally emerge on the 4th and 5th days after contact [11]. The most common
symptoms are fever, cough, fatigue, and shortness of breath. The diagnostic
approaches for corona virus diagnosis are basic molecular experiments on respira-
tory samples (esophagus swab/nasopharyngeal swab/saliva/bronchoalveolar lavage
and endotracheal extracts). SARS-CoV-2 RNA, determined by reverse transcription
polymerase chain reaction, is collected from the nasopharyngeal swab widely.
Apart from the nasopharyngeal swab, virus particles are also detected in the stool
and blood. Other laboratory examinations are not specific to the virus and indicate
infection. In the measurements made, the white-cell count is normal or low. The
platelet count in the whole blood test is usually normal. C-reactive protein (CRP) is
Natural adjunctive therapies options other than COVID-19 407
elevated due to inflammation. Procalcitonin levels are normal and elevated, which
should suggest a bacterial coinfection. Elevated prothrombin time, creatinine, and
D-dimer levels indicate the severity of the disease. Chest CT scan presented as an
indicator for pneumonia is important clinical assay. SARS-CoV-19 virus pre-
dominantly affects the pulmonary system. First, cells in the respiratory system are
infected by SARS-CoV-19 and cause inflammation and are observed with some
clinical symptom. In pulmonary involvement, a diffuse ground glass appearance
is seen on chest tomography. However, the mechanism of the infection pathway is
not clear. Angiotensin-converting enzyme 2 (ACE2) receptor for entry into the
cells that are abundant in the lower respiratory tract. Importantly, in the brain,
heart, intestinal epithelium, vascular endothelium, and the kidneys is expressed
ACE2 that makes potential targets capacity for all these organs. In addition, other
organ involvement, septic shock, and metabolic acidosis may occur [12].
According to the pathological findings of autopsies and biopsies restriction, in
addition to the lung SARS-CoV-19 can inter numerous tissues and organs, as the
liver, spleen, heart, brain and kidney [13].
15.2 Immune system and inflammatory responds

Although the physiopathogenesis of SARS-CoV-2 in the body is not known
exactly, it is similar to other coronavirus family. SARS-CoV-2 first binds to epi-
thelial cells in the upper respiratory tract, namely the oral and nasal cavities. It then
proceeds to the lower respiratory tract and infects alveolar type II pneumocytes.
SARS-CoV-2 invades cells using ACE2 receptors [14]. ACE2 is an amino pepti-
dase enzyme that changes angiotensin (Ang) II into Ang (1-7). It is well known that
the function of Ang II is a major role in contraction blood vessel, proinflammatory,
and pro-fibrotic effects. In contrast, Ang (1-7) is a potent vasodilator, an inhibitor
of the apoptotic, and an inhibitor of the proliferative agent. In almost all of the
pathological conditions, especially those of the cardiovascular system, ACE
enzyme prevents ACE2 enzyme in your body that produces Ang II. Furthermore,
the two enzymes ACE/ACE2 play a major role in the renin–angiotensin system
homeostasis that regulates blood pressure. In humans, ACE2 receptors are found in
many organs. These can be listed as upper respiratory system (nasopharynx, nasal,
and oral mucosa), lower respiratory system epithelium, gastrointestinal system
(small intestine, colon), kidney, liver, and vascular endothelial cells [15]. The S
protein recognizes the ACE2 receptor in the body and provides entry into the cell
[16]. After the virus enters the cell, the viral RNA is first copied. Then membrane
proteins (S, M, and E) are synthesized. It is transported to the endoplasmic reti-
culum–Golgi for packaging. Together with genomic RNA, it forms N protein
nucleocapsids. The pathogen is transported to the plasma membrane and excreted
by exocytosis [17]. SARS-CoV-2 binds to ACE2, causing downregulation of ACE2
receptors, leading to an increase in Ang II. Ang II causes pulmonary vasocon-
striction, acute lung injury, and pulmonary edema and fibrosis [18]. Endothelial
dysfunction and hypercoagulation may result in pulmonary embolism in
COVID-19 [19]. COVID-19 tries to destroy the respiratory tract and begins to
grow quickly in the early stages of infection, though many patients remain
asymptomatic [20–22]. Monocytes and macrophages show the first response. Viral
replication continues. Vasodilation, endothelial permeability, and leukocyte
aggregation are all caused by tissue injury and chronic inflammation. The chronic
inflammation in some individuals continues to escalate, leading in systemic
inflammation. Several distal organs can be harmed by systemic problems. COVID-
19 infection causes cytokine storm and multiple organ damage. The occurrence of
cytokine storm is associated with age and gender. In the cytokine storm, proin-
flammatory cytokines are released uncontrollably and their amounts increase.
Excessive cytokine release that occurs causes acute respiratory distress syndrome
(ARDS) development in the lungs [23]. When the virus enters macrophages and
dendritic cells, it leads to the activation of immunity [24]. It causes cytokine release
after CD4þ T cells. Uncontrolled release of cytokines and mediators causes cyto-
kine storm syndrome. The exact mechanism of this syndrome against viral infec-
tions is still unknown [25]. Acute lung injury is the result of cytokine storm
syndrome. This is seen in patients with SARS-CoV-2 infection, lung disease that
can lead to widespread lung damage, inflammation, and ultimately death. High
amounts of proinflammatory cytokines and chemokines have been found in the
plasma of patients infected with SARS-CoV-2. It has been found that patients with
critical illness have higher cytokine concentrations in their plasma [12]. Interleukin
6 (IL-6) plays an important role in cytokine storm syndrome. It has both anti-
inflammatory and proinflammatory effects. IL-6 levels have been shown to be
associated with the severity of the disease [26]. Whether SARS-CoV-2 releases
immunity, and if so, which antibody titers are protective is still a matter of debate. It
has been observed that infected patients develop antibody titers 10–15 days after
symptom onset. More research is needed to see if antibody titers can predict whe-
ther or not a person will recover first from condition.
15.3 Proinflammatory cytokines

During illness in serum IL-1B, IL-6, interferon-gamma (IFNg), IFNg-inducible
protein-10 (IP10) and monocyte chemotactic protein-1 (MCP1) increase levels of
certain inflammatory cytokines. Increasing the amount of proinflammatory cyto-
kine increases the number and function of helper type 1 (Th1) lymphocytes. IL-4
and IL-10 with increasing T-helper type 2 (Th2) are the activation of lymphocytes.
At high levels in people is severe disease granulocyte colony stimulating factor,
IP10, MCP1, macrophage inflammatory protein 1, and TNFa was determined [9].
In COVID-19, a “cytokine storm” is an important condition that determines the
severity of the disease. In the cytokine storm, a large number of immune cells
arrive at the infection site, the vascular barrier is damaged, and as a result, multiple
organ failure occurs. COVID-19 has been observed to increase the levels of many
cytokines. IL-1, TNFa, and IL-6 are the most important proinflammatory cyto-
kines. Increase in IL-6 is associated with higher mortality [27].
15.4 Immunomodulators and adjunctive therapies

The best protection from COVID-10 is to observe social distance, masks, or per-
sonal hygiene. However, a healthy lifestyle, exercise, and a healthy diet are also
important. Unfortunately, there is no gold standard treatment protocol for COVID-
19. This global epidemic must be fought with all available data, including artificial
intelligence [28]. Although supporting protocols are recommended, a clear protocol
has not been established. Unfortunately, there is currently no vaccine or medical
treatment with proven efficacy against SARS-CoV-2 infection. All current studies
are conducted with drugs that have previously been shown to be effective in the
treatment of SARS-CoV. There is no drug that will directly affect the patient,
which will directly affect the virus. For this reason, adjunctive therapies, whose
attachments are linked to their previous viral effects, are very valuable. In the
treatment of COVID-19, it is necessary to increase immunity and reduce inflam-
mation. In this, it is necessary to reduce the virus–cell relationship and prevent
replication. It has been observed that immunomodulators and adjunctive therapies
are used to give benefit to COVID-19 patients through antiviral, anti-inflammatory,
antioxidant, antithrombotic and immunomodulatory effects.
15.4.1 Phenolic compounds

Phenolic compounds are widely distributed in plants. These compounds are a large
group of molecules that have various functions in plant growth, development, and
defense, as seconder metabolites in the plants. The compounds include those
compounds that can protect the plant against insects, fungi, bacteria, and viruses.
Also, there are signal molecules, pigments, and flavors that can attract or repel in
phenolic compounds. During this century, there was major changing in trends of
consumer preferences toward the natural products owing to the growing knowledge
about the health ailment arising due to synthetic additives. Phenolic compounds
found in plants are considered to be one of the main stones of the human diet. Many
of these phenolic compounds found in natural foods have antioxidant properties.
While these compounds are not essential for plant growth, they are essential for
human health. Many studies have been conducted on phenolic compounds due to
their antioxidant effects.
Phenolic compounds are widely found in nature and have biological and
pharmacological activities. There is an aromatic ring attached to the hydroxyl
group in their structure. The structure of phenolic compounds varies widely from
simple to complex. Phenolic compounds play an important role in the food chain
due to their antioxidant effects. These compounds with antioxidant effects are used
to extend the shelf life of foods. Unlike synthetic antioxidants, being natural pro-
vides additional gain.
There is increasing awareness that these bioactive compounds can have sig-
nificant positive effects on people’s health, physical well-being, and mental states.
Prevention and treatment of human diseases in different countries is possible by
herbs and plant extracts for medical purposes. The effects of these compounds
Table 15.1 Different classes of flavonoids and dietary sources
Class Name Dietary source

Flavones Chrysin Fruit skins
Apigenin Parsley, celery
Flavonones Naringin Citrus, grapefruits
Eriodictyol Lemons
Hesperidin Oranges
Flavonols Quercetin Onion, lettuce, tea, berries, broccoli,
tomato, apples, olive oil, cranberry
Rutin Buckwheat, tomato skin, citrus, red
pepper, red wine
Flavononols Astilbin White grape skin
Genistin Soybeans
Taxifolin Fruits
Isoflavones Daidzin Soybeans
Daidzein Soybeans
Genistein Soybeans
Flavanols (þ)-Catechin Tea
(þ)-Gallocatechin Tea
()-Epicatechin Tea
()-Epigallocatechin Tea
()-Epicatechin gallate Tea
()-Epigallocatechin gallate Tea
Anthocyanidins Cyanidin Cherry, grapes, strawberry, raspberry
Epigenidin Stored fruits
Delphinium Dark fruits
Pelargonidin Dark fruits
present in plants show that extensive research is required for drug production.
Naturally, occurring antiviral compounds with least toxicity have been found to be
the best choice against coronaviruses. Many bioactive compounds are produced as
secondary metabolites in plants. The most useful classification of phenolics in two
major companies, flavonoids and non-flavonoid polyphenols, is shown in
Table 15.1. There are over 6,000 different flavonoids that have been identified, and
this number continues to increase. Flavonoids, a large group of phenolic com-
pounds, are found in many fruits and vegetables, giving color and flavor to the
foods they are found in. The antioxidant effects of flavonoids are for having three
properties. Metal chelating capability, which is significantly influenced by the
configuration of hydroxyl and carbonyl groups surrounding the molecule, the
availability of hydrogen that is capable of reducing free radicals, and the flavo-
noid’s capacity to delocalize the unpaired electron precursors are all important
factors. Flavonoids show their antioxidant effects by chain-breaking reactions. Due
to the synergistic effects with other products in plants, instead of taking phenolic
compounds as supplements, taking them naturally into foods makes their anti-
oxidant effects stronger. Recently, antioxidants have gained importance due to their
role in free radical scavenging to stop the process of oxidation. Nowadays, natural
antioxidants importance in human health and nutrition is increased by researching
of antioxidants. The Mother Nature’s defender can boost up the immune system,
lower cholesterol and blood pressure level, many benefits from a compound that
comes from many fruits and vegetables. Polyphenolic compounds such as flavo-
noids have been used since ancient times for their positive effects on health, even
before modern medicine was developed. The phenolics are fascinating biologically
active compounds that have exhibited great promise in both the inhibition and
treatment of many human ailments. Plants create these chemicals in responses to
damage or infection by microbial or fungal pathogens. It has antioxidant, anti-
inflammatory, antitumor, antimicrobial effects. Biologically, active compounds act
by targeting specific sites in cells. It has direct effects on proteins, receptors, and
enzymes. They affect the inhibition or activation of enzymes. They provide a
modification of proteins. Phenolic compounds are attached to peptides and proteins
by hydrogen bonding. As a group effect, it is valuable for human health due to their
antioxidation, anti-inflammation, modulation of signal transduction, anti-microbial
activity, and anti-proliferation activities. Use of phytochemicals as antiviral agents
provides a hope against proliferation of SARS-CoV-2.
15.4.1.1 Resveratrol
Polyphenols, nonflavonoids, and flavonoids are abundant in red wine, grapes,
berries, and nuts. These compounds have been shown to have antioxidant,
antitumor, and antiviral effects in the organism [29]. Resveratrol (RESV) is a
nonflavonoid compound and has anti-inflammatory properties [30]. Besides its
anti-inflammatory effect, RESV contributes to antiviral and cellular survival.
RESV has been shown to disrupt the SARS-CoV-2 spike protein and to inhibit
ACE2 receptor binding [31]. Despite these effects, many studies are needed for
effective dosage and safe use.
15.4.1.2 Celastrol
Celastrol has been shown to have anti-inflammatory effects in lung diseases by
suppressing NF-kB signaling [32]. It has also been shown to have curative effects
in lung diseases by increasing antioxidant defenses by decreasing levels of IL-8,
TNFa, and monocyte chemoattractant protein-1 [33]. It has been shown to control
ARDS by reducing proinflammatory cytokines and NF-kB activation [34].
Celastrol has been shown to decrease S protein proliferation and cell viral entry
[35].
15.4.1.3 Curcumin
Curcumin is a phenolic acid that is found in turmeric. Curcumin possesses anti-
inflammatory and immunomodulatory properties, analogous to many other plant
phenolic acids [36]. Curcumin protects the cell entrance by binding to the SARS-
CoV-2 virus’s S protein and the ACE2 receptors protein’s viral binding ability [37].
It is known that curcumin’s antithrombotic, anti-cytokine, and antifibrotic proper-
ties can be used in COVID-19 patients [38]. Despite the known effects of curcumin,
more studies are needed because the safety dose in COVID-19 patients is
not known.
15.4.1.4 Quercetin
Quercetin may be found in abundance in fruits like apples and citrus, as well as
vegetables like onion, broccoli, and tomato. Quercetin functions as an antioxidant
that protects bodily tissues from oxidative stress and boosts cell survival rates [39].
Quercetin has high anti-inflammatory activity. It does this by inhibiting the
cyclooxygenase enzyme and by inhibiting prostaglandin synthesis [40]. Quercetin
decreases viral replication, resulting in decreased viral load.
15.4.1.5 Bee products

Interest in alternative medicine to bee products has increased significantly over the
last decade [41]. Propolis has been used from ancient ages due to its useful features in
the popular medicine for the treat of various diseases. Most investigators have sup-
ported the pharmacological mechanisms as antiseptic, anti-inflammatory, antioxidant
features of propolis with their studies [42,43]. Propolis is a gum made by bees from
bioactive plant parts. It is currently used as adjuvant therapy in diseases. Anti-
inflammatory, immunomodulatory, antitumoral, and antioxidant effects have been
proven [44]. While there are many studies proving the antimicrobial activity of
propolis, there are limited studies for its antiviral activity. However, it should not be
forgotten that bee products contain plenty of phenolic compounds and have
predictable antiviral, immune modulatory, and antioxidant effects. The chemical
composition of propolis contains many functional compounds. One of the most
effective components of propolis is caffeic acid phenethyl ester that is a flavonoid-
like compound. Caffeic acid phenethyl ester is an antioxidant component. Pollen, as
another honey bee product, is a source of protein, amino acid, mineral, and vitamin
for human health. Pollen is collected from many plants by the honey bee. Pollen is a
powerful antioxidant and is a polyphenolic compound [45–47]. Phenolic compounds
are found in propolis and pollen structure. Due to its antioxidant effects, NF-kB
(nuclear factor-kB) regulates the expression and transcription factors of adhesion
molecules in endothelial cells. NF-kB is activated with an increase in reactive oxy-
gen species. Anti-inflammatory activities of phenolics may be mediated by tran-
scriptional factor NF-kB in cells [48]. Bee products are known to be potent
antioxidants. It was observed that NF-kB activity increased when there was oxidative
damage [49–51]. NF-kB is an oxidative stress response factor. Antioxidants can
block activate NF-kB [52]. Since there is no clear treatment protocol defined for
COVID-19, all therapies that we think can be effective are very valuable.
15.4.2 Melatonin
Melatonin, a pineal gland hormone, has anti-inflammatory and immunomodulatory
effects. Melatonin modulates cytokine release by affecting T and B cells with mel-
atonin receptors [53]. One reason for the severe course of COVID-19 in the elderly
population may be associated with low melatonin levels [54]. An amount of 2 mg of
melatonin taken in the evening is healthy. Administration of COVID-19 melatonin

should begin with the contamination of the individual with the sick person.
15.4.3 Zinc
The trace mineral zinc plays an important role in the inflammatory process.
Proinflammatory cytokines increase in zinc deficiency. In zinc deficiency, proin-
flammatory cytokines sensitivity increases in lung epithelial tissues; in addition,
cell-barrier functions are impaired. This is why zinc is an essential element in
COVID-19 infection [55]. Zinc increases the amount of IL-2, an anti-inflammatory
cytokine. In addition, it inhibits the proliferation of coronavirus [56].
15.4.4 Ascorbic acid (vitamin C)

As is known, ascorbic acid in vitamin C is a powerful antioxidant and immune
enhancing, especially in high doses. Its powerful antioxidant effect can be used as
adjunctive therapy in symptomatic COVID-19 patients to clear cellular toxicity.
Vitamin C is effective in infection control. In a study, it was observed that cold
attacks decreased with vitamin C support [57]. At the same time, vitamin C has
been shown to have an effect on CRP levels [58]. In a study, considering the known
benefits of vitamin C, it was applied to COVID-19 patients and a significant
decrease in inflammatory markers was detected in patients with severe COVID-19
disease [59]. TNF- and other proinflammatory cytokines are known to be reduced
by vitamin C, whereas anti-inflammatory cytokines are increased (IL-10). It has
been shown that vitamin C increases the secretion of IL-10. IL-10 operates on the
basis of a negative feedback loop with IL-6 so that vitamin C indirectly controls the
inflammatory cytokine IL-6 by controlling IL-10 levels [60]. Administration of
vitamin C can positively affect COVID-19 patients with ARDS by reducing
inflammation, pathogen infectiousness, and virulence, as well as optimizing the
immune defense [61]. Cytokine storm, which is a feared situation in patients with
COVID-19, can be controlled with vitamin C.
15.4.5 Vitamin D
Vitamin D is a steroid hormone precursor that is fat-soluble. Vitamin D deficiency
is a common health problem, and there are many factors that cause it. In the
COVID-19 pandemic, vitamin D deficiency was frequently observed in severely ill
patients, suggesting the importance of vitamin D supplementation [10]. Vitamin D
enhances cellular immunity by increasing the expression of antimicrobial peptides.
Vitamin D is known to increase intercellular connections, thereby reducing pul-
monary edema caused by viruses [62]. Vitamin D regulates the immune response
by modulating the release of proinflammatory cytokines while increasing the
release of anti-inflammatory cytokines [63]. Although this has been claimed that
vitamin D insufficiency enhances the incidence and severity of COVID-19 infes-
tation, there has been no definitive research to support this theory. Vitamin D intake
is indicated to increase immunity versus COVID-19 and prevent human fatalities,
according to the theory. Vitamin D deficiency is considered to help defend the
respiratory epithelium against bacterial spread [64]. RP, a marker of inflammation,

has been correlated with vitamin D deficiency [65]. Vitamin D supplementation
increases ACE2 concentrations by decreasing RAS activity in acute lung injury
[66]. An increase in ACE2 leads to a decrease in viral activation. Vitamin D levels
have been found to be low in severe COVID-19 patients. Vitamin D replacement is
still a matter of debate [67].
15.4.6 Vitamin E
Vitamin E is the main component of antioxidant defense. Vitamin E aids immune
regulation by regulating immune cells and increasing IL-2 cytokine secretion
through its antioxidant effect [68]. In the studies, the protective effect of vitamin E
from respiratory tract infections was mentioned. Both immunomodulator and
antioxidant effects are important for COVID-19 so vitamin E is one of the essential
vitamins for adjunctive therapies.
15.4.7 Selenium
Selenium (Se) is a vital micronutrient for public health. Se deficiency is a common
health condition worldwide. Se viral infections are known to be the determinant of
the response. Although Se levels are high in seafood, for most foods, the Se content
depends on the Se level in the soil. The trace element selenium is the main com-
ponents of the antioxidant defense. Studies have shown that the immune response
changes in Se deficiency. A study conducted in China found that the change in Se
levels changed the rate of recovery from COVID-19 [69]. It is important to prevent
cytokine storm, which is one of the most important causes of mortality in COVID-
19 patients; therefore elements with high antioxidant and immunomodulatory
effects such as Se are important [70].
15.4.8 Omega-3 fatty acids

Omega-3 fatty acids are polyunsaturated fatty acids. Severe complications of
COVID-19 are seen as a result of cytokine storm. It occurs due to intravascular
disseminated coagulation, ARDS, and multiple organ dysfunction syndrome,
resulting in death from cytokine storm. For this reason, substances with anti-
inflammatory effects are important. They can control the cytokine storm due to the
anti-inflammatory, immunomodulatory, and antiviral effects of eicosapentaenoic
acid and docosahexaenoic acid, which are omega-3 fatty acids. Omega-3 fatty acids
activate macrophages by increasing cytokine release [71]. It reduces the inflam-
matory process by downregulating the nuclear factor-beta of omega-3 fatty acids.
Among lipids, omega-3 fatty acids can inactivate viruses by regulating lipid status.
On the other hand, omega-3 fatty acids can help suppress prostaglandin (proin-
flammatory) production by inhibiting cyclooxygenase enzymes. It also reduces the
platelet-activating factor. Thromboembolic events are common in COVID-19, and
reducing or inhibiting platelet activation can reduce these events [72]. Because of
these effects, omega-3 fatty acids are important in adjunctive natural therapies in
COVID-19.
15.5 Dietary ingredients in immunity

There is limited information on nutritional support for COVID-19 patients. However,
nutritional therapy has a negligible effect. Foods that have a role on the immune
system in patients diagnosed with COVID-19 play an important role. Natural
bioactive compounds are valuable in COVID-19 patients for their anti-inflammatory
effects and their effects on controlling the disease. As discussed, many natural pro-
ducts can reduce the symptom burden of COVID-19. They have also been shown to
have antiviral properties and boost immunological responses. Organic compounds of
these minerals and vitamins include fruits, vegetables, meats, fish, chicken, and milk
products. Many studies on the effect of nutrition on the immune system have been
conducted from past to present. The relationship between COVID-19 and nutrition
has just begun to be investigated. The quality and quantity of proteins are an
important basis for the immune system. Due to their anti-inflammatory effects, foods
containing essential amino acids are in the food group of high biological value. We
can sample eggs, meat, fish, and milk [73]. The question is whether dietary vitamins
and minerals are adequate for COVID-19. For these substances, the effective dose
and duration of use are still unknown.
15.6 Conclusion and future scope for natural antiviral

therapies against COVID-19
The distribution of the COVID-19 outbreak differs in all geographic regions.
Although the reason for this is not known clearly, it can be associated with eating
habits, use of natural medicines, vitamins, or minerals. Since there is no medical
antiviral treatment with a net effect, adjunctive therapies come to light. Although
there is no clear evidence with these therapies, studies should be increased on these
therapies that slow viral spread and increase the response to treatment. When
contamination with COVID-19 occurs, it is necessary to protect from the disease by
keeping the immune system strong or by reducing its inflammatory reactions in the
case of disease. Although there is no mono therapies affecting these conditions,
many studies are needed for the use of mono or combined therapies.
References
[1] Wang D., Hu B., Hu C., et al. ‘Clinical characteristics of 138 hospitalized
patients with 2019 novel coronavirus-infected pneumonia in Wuhan, China’.
JAMA. 2020;323(11):1061–9.
[2] Muhammad L.J., Algehyne E.A., Usman S.S., Ahmad A., Chakraborty C.
and Mohammed I.A. ‘Supervised machine learning models for prediction of
2021;2(11):1–13, 2.
[3] Chakraborty C., Gupta B. and Ghosh S.K. ‘Mobile metadata assisted com-
munity database of chronic wound images’. Wound Medicine. 2014;6:34–42.
[4] Zhu N., Zhang D., Wang W., et al. China Novel Coronavirus Investigating
and Research Team. ‘A novel coronavirus from patients with pneumonia in
China, 2019’. New England Journal of Medicine. 2020;382(8):727–33.
[5] Zheng J. ‘SARS-CoV-2: An emerging coronavirus that causes a global
threat’. International Journal of Biological Sciences. 2020;16(10):1678–85.
[6] Velavan T.P. and Meyer C.G. ‘The COVID-19 epidemic’. Tropical
Medicine & International Health. 2020;25(3):278–80.
[7] Masters P.S. ‘The molecular biology of coronaviruses’. Advances in Virus
Research. 2006;66:193–292.
[8] Arons M.M., Hatfield K.M., Reddy S.C., et al.; Public Health–Seattle and
King County and CDC COVID-19 Investigation Team. ‘Presymptomatic
SARS-CoV-2 infections and transmission in a skilled nursing facility’. The
New England Journal of Medicine. 2020;382(22):2081–90.
model’. IEEE Access. 2020;8:159402–14.
[10] Pan X., Chen D., Xia Y., et al. ‘Asymptomatic cases in a family cluster with
SARS-CoV-2 infection’. The Lancet. Infectious Diseases. 2020;20(4):410–1.
[11] Guan W.J., Ni Z.Y., Hu Y., et al.; China Medical Treatment Expert Group
for Covid-19. ‘Clinical characteristics of coronavirus disease 2019 in China’.
The New England Journal of Medicine. 2020;382(18):1708–20.
[12] Huang C., Wang Y., Li X., et al. ‘Clinical features of patients infected with
2019 novel coronavirus in Wuhan, China’. The Lancet. 2020;395
(10223):497–506.
[13] Xu Z., Shi L., Wang Y., et al. ‘Pathological findings of COVID-19 asso-
ciated with acute respiratory distress syndrome’. The Lancet. Respiratory
Medicine. 2020;8(4):420–2.
[14] Zhou P., Yang X.L., Wang X.G., et al. ‘A pneumonia outbreak associated
with a new coronavirus of probable bat origin’. Nature. 2020;579
(7798):270–3.
[15] Hamming I., Timens W., Bulthuis M.L., Lely A.T., Navis G. and van Goor
H. ‘Tissue distribution of ACE2 protein, the functional receptor for SARS
coronavirus. A first step in understanding SARS pathogenesis’. The Journal
of Pathology. 2004;203(2):631–7.
[16] Shereen M.A., Khan S., Kazmi A., Bashir N. and Siddique R. ‘COVID-
19 infection: Origin, transmission, and characteristics of human cor-
onaviruses’. Journal of Advanced Research. 2020;24:91–8.
[17] Li X., Geng M., Peng Y., Meng L. and Lu S. ‘Molecular immune patho-
genesis and diagnosis of COVID-19’. Journal of Pharmaceutical Analysis.
2020;10(2):102–8.
[18] Imai Y., Kuba K., Rao S., et al. ‘Angiotensin-converting enzyme 2 protects
from severe acute lung failure’. Nature. 2005;436(7047):112–6.
[19] Wang M., Hao H., Leeper N.J. and Zhu L.; Early Career Committee.
‘Thrombotic regulation from the endothelial cell perspectives’. Arteriosclerosis,
Thrombosis, and Vascular Biology. 2018;38(6):e90–5.
[20] Bhuyan H.K., Chakraborty C., Shelke Y. and Pani S.K. ‘COVID-19 diag-
nosis system by deep learning approaches’. Expert Systems. 2021: e12776.
[21] Dash S., Chakraborty C., Giri S.K., Pani S.K. and Frnda J. ‘BIFM: Big-data
driven intelligent forecasting model for COVID-19’. IEEE Access. 2021:1–
13. 10.1109/ACCESS.2021.3094658.
[22] Aslam B., Javed A.R., Chakraborty C., Nebhen J., Raqib S. and Rizwan M.
Computing. 2021:1–17. 10.1007/s00779-021-01596-3.
[23] Coperchini F., Chiovato L., Croce L., Magri F. and Rotondi M. ‘The cyto-
kine storm in COVID-19: An overview of the involvement of the chemo-
kine/chemokine-receptor system’. Cytokine & Growth Factor Reviews.
2020;53:25–32.
[24] Li G., Fan Y., Lai Y., et al. ‘Coronavirus infections and immune responses’.
Journal of Medical Virology. 2020;92(4):424–32.
[25] Tisoncik J.R., Korth M.J., Simmons C.P., Farrar J., Martin T.R. and Katze
M.G. ‘Into the eye of the cytokine storm’. Microbiology and Molecular
Biology Reviews. 2012;76(1):16–32.
[26] Coomes E.A. and Haghbayan H. ‘Interleukin-6 in Covid-19: A systematic
review and meta-analysis’. Reviews in Medical Virology. 2020;30(6):1–9.
[27] Ragab D., Salah Eldin H., Taeimah M., Khattab R. and Salem R. ‘The
COVID-19 cytokine storm; what we know so far’. Frontiers in Immunology.
2020;11:1446.
[28] Chakraborty C. and Abougreen A.N. ‘Intelligent Internet of Things and
Transactions on Pervasive Health and Technology. 2021:1–14.
[29] Bravo L. ‘Polyphenols: Chemistry, dietary sources, metabolism, and nutri-
tional significance’. Nutrition Reviews. 1998;56(11):317–33.
[30] Gao Y., Fu R., Wang J., Yang X., Wen L. and Feng J. ‘Resveratrol mitigates
the oxidative stress mediated by hypoxic-ischemic brain injury in neonatal
rats via Nrf2/HO-1 pathway’. Pharmaceutical Biology. 2018;56(1):440–9.
[31] Wahedi H.M., Ahmad S. and Abbasi S.W. ‘Stilbene-based natural com-
pounds as promising drug candidates against COVID-19’. Journal of
Biomolecular Structure & Dynamics. 2021;39(9):3225–34.
[32] Habtemariam S., Nabavi S.F., Berindan-Neagoe I., et al. ‘Should we try the
antiinflammatory natural product, celastrol, for COVID-19?’. Phytotherapy
Research. 2020;34(6):1189–90.
[33] Shi K., Chen X., Xie B., et al. ‘Celastrol alleviates chronic obstructive pul-
monary disease by inhibiting cellular inflammation induced by cigarette
smoke via the Ednrb/Kng1 signaling pathway’. Frontiers in Pharmacology.
2018;9:1276.
[34] Wei Y. and Wang Y. ‘Celastrol attenuates impairments associated with
lipopolysaccharide-induced acute respiratory distress syndrome (ARDS) in
rats’. Journal of Immunotoxicology. 2017;14(1):228–34.
[35] Iwata-Yoshikawa N., Okamura T., Shimizu Y., Hasegawa H., Takeda M.
and Nagata N. ‘TMPRSS2 contributes to virus spread and immunopathology
in the airways of murine models after coronavirus infection’. Journal of
Virology. 2019;93(6):e01815–8.
[36] Aggarwal B.B., Gupta S.C. and Sung B. ‘Curcumin: An orally bioavailable
blocker of TNF and other pro-inflammatory biomarkers’. British Journal of
Pharmacology. 2013;169(8):1672–92.
[37] Das S., Sarmah S., Lyndem S. and Singha Roy A. ‘An investigation into the
identification of potential inhibitors of SARS-CoV-2 main protease using
molecular docking study’. Journal of Biomolecular Structure & Dynamics.
2021;39(9):3347–57.
[38] Lelli D., Sahebkar A., Johnston T.P. and Pedone C. ‘Curcumin use in pul-
monary diseases: State of the art and future perspectives’. Pharmacological
Research. 2017;115:133–48.
[39] Xu D., Hu M.J., Wang Y.Q. and Cui Y.L. ‘Antioxidant activities of quercetin
and its complexes for medicinal application’. Molecules. 2019;24(6):1123.
[40] Jan A.T., Kamli M.R., Murtaza I., Singh J.B., Ali A. and Haq Q.M.R.
‘Dietary flavonoid quercetin and associated health benefits – An overview’.
Food Reviews International. 2010;26(3):302–17.
[41] Pasupuleti V.R., Sammugam L., Ramesh N. and Gan S.H. ‘Honey, propolis,
and royal jelly: A comprehensive review of their biological actions and
health benefits’. Oxidative Medicine and Cellular Longevity. 2017;2017:
1259510.
[42] Banskota A.H., Tezuka Y. and Kadota S. ‘Recent progress in pharmacolo-
gical research of propolis’. Phytotherapy Research. 2001;15(7):561–71.
[43] Burdock G.A. ‘Review of the biological properties and toxicity of bee pro-
polis (propolis)’. Food and Chemical Toxicology. 1998;36(4):347–63.
[44] Sforcin J.M. ‘Biological properties and therapeutic applications of propolis’.
Phytotherapy Research. 2016;30(6):894–905.
[45] Mărghitas L.A., Stanciu O.G., Dezmirean D.S., et al. ‘In vitro antioxidant
capacity of honeybee-collected pollen of selected floral origin harvested
from Romania’. Food Chemistry. 2009;115(3):878–83.
[46] Talas Z.S., Ozdemir I., Ciftci O., Cakir O., Gulhan M.F. and Pasaoglu O.M.
‘Role of propolis on biochemical parameters in kidney and heart tissues
against L-NAME induced oxidative injury in rats’. Clinical and
Experimental Hypertension. 2014;36(7):492–6.
[47] Selamoglu Z.S., Ozdemir I., Ciftci O., Gulhan M.F. and Savci A.
‘Antioxidant effect of ethanolic extract of propolis in liver of L-NAME
treated rats’. Advances in Clinical and Experimental Medicine. 2015;24
(2):227–32.
[48] Babu P.V. and Liu D. ‘Green tea catechins and cardiovascular health: An
update’. Current Medicinal Chemistry. 2008;15(18):1840–50.
[49] Eckerson H.W., Wyte C.M., and La Du B.N. ‘The human serum para-
oxonase/arylesterase polymorphism’. American Journal of Human Genetics.
1983;35(6):1126–38.
[50] Zheng J.P., Zhang Y., Edvinsson L., Hjalt T. and Xu C.B. ‘NF-kappaB
signaling mediates vascular smooth muscle endothelin type B receptor
expression in resistance arteries’. European Journal of Pharmacology.
2010;637(1–3):148–54.
[51] Simão S., Gomes P., Pinto V., et al. ‘Age-related changes in renal expression
of oxidant and antioxidant enzymes and oxidative stress markers in male
SHR and WKY rats’. Experimental Gerontology. 2011;46(6):468–74.
[52] Xu J., Fan G., Chen S., Wu Y., Xu X.M. and Hsu C.Y.
‘Methylprednisolone inhibition of TNF-alpha expression and NF-kB acti-
vation after spinal cord injury in rats’. Brain Research. Molecular Brain
Research. 1998;59(2):135–42.
[53] Srinivasan V., Maestroni G.J., Cardinali D.P., Esquifino A.I., Perumal S.R.
and Miller S.C. ‘Melatonin, immune function and aging’. Immunity &
Ageing. 2005;2:17.
[54] Gunn P.J., Middleton B., Davies S.K., Revell V.L. and Skene D.J. ‘Sex
differences in the circadian profiles of melatonin and cortisol in plasma and
urine matrices under constant routine conditions’. Chronobiology
International. 2016;33(1):39–50.
[55] Skalny A.V., Rink L., Ajsuvakova O.P., et al. ‘Zinc and respiratory tract
infections: Perspectives for COVID-19 (Review)’. International Journal of
Molecular Medicine. 2020;46(1):17–26.
[56] te Velthuis A.J., van den Worm S.H., Sims A.C., et al. ‘Zn(2þ) inhibits
coronavirus and arterivirus RNA polymerase activity in vitro and zinc
ionophores block the replication of these viruses in cell culture’. PLoS
Pathogens. 2010;6(11):e1001176.
[57] Johnston C.S., Barkyoumb G.M. and Schumacher S.S. ‘Vitamin C supple-
mentation slightly improves physical activity levels and reduces cold inci-
dence in men with marginal vitamin C status: A randomized controlled trial’.
Nutrients. 2014;6(7):2572–83.
[58] Jafarnejad S., Boccardi V., Hosseini B., Taghizadeh M. and Hamedifard Z.
‘A meta-analysis of randomized control trials: The impact of vitamin C
supplementation on serum CRP and serum hs-CRP concentrations’. Current
Pharmaceutical Design. 2018;24(30):3520–8.
[59] Hiedra R., Lo K.B., Elbashabsheh M., et al. ‘The use of IV vitamin C for
patients with COVID-19: A case series’. Expert Review of Anti-Infective
Therapy. 2020;18(12):1259–61.
[60] Zhou L. and Xu H. ‘A Clinical Study for the Efficacy and Safety of
Adalimumab Injection in the Treatment of Patients With Severe Novel
Coronavirus Pneumonia (COVID-19)’. 2020. http://www.chictr.org.cn/
showprojen.aspx?proj¼49889.
[61] Hoang B.X., Shaw G., Fang W. and Han B. ‘Possible application of high-
dose vitamin C in the prevention and therapy of coronavirus infection’.
Journal of Global Antimicrobial Resistance. 2020;23:256–62.
[62] Gorman S., Buckley A.G., Ling K.M., et al. ‘Vitamin D supplementation of
initially vitamin D-deficient mice diminishes lung inflammation with limited
effects on pulmonary epithelial integrity’. Physiological Reports. 2017;5

(15):e13371.
[63] Jeffery L.E., Burke F., Mura M., et al. ‘1,25-Dihydroxyvitamin D3 and IL-2
combine to inhibit T cell production of inflammatory cytokines and promote
development of regulatory T cells expressing CTLA-4 and FoxP3’. Journal
of Immunology. 2009;183(9):5458–67.
[64] Merzon E., Tworowski D., Gorohovski A., et al. ‘Low plasma 25(OH)
vitamin D level is associated with increased risk of COVID-19 infection: An
Israeli population-based study’. The FEBS Journal. 2020;287(17):3693–702.
[65] Daneshkhah A., Agrawal V., Eshein A., Subramanian H., Roy H.K. and
Backman V. ‘Evidence for possible association of vitamin D status with
cytokine storm and unregulated inflammation in COVID-19 patients’. Aging
Clinical and Experimental Research. 2020;32(10):2141–58.
[66] Xu J., Yang J., Chen J., Luo Q., Zhang Q. and Zhang H. ‘Vitamin D alleviates
lipopolysaccharide-induced acute lung injury via regulation of the renin-
angiotensin system’. Molecular Medicine Reports. 2017;16(5):7432–8.
[67] Panagiotou G., Tee S.A., Ihsan Y., et al. ‘Original publication: Low serum
25-hydroxyvitamin D (25[OH]D) levels in patients hospitalized with
COVID-19 are associated with greater disease severity’. Clinical
Endocrinology. 2020;93(5):629–30.
[68] Wu D. and Meydani S.N. ‘Vitamin E, immune function, and protection
against infection’. In: Weber P., Birringer M.,Blumberg J., Eggersdorfer M.,
and Frank J. (eds.) Vitamin E in Human Health. Nutrition and Health.
(pp. 371–384). Cham: Springer; 2019.
[69] Zhang J., Taylor E.W., Bennett K., Saad R. and Rayman M.P. ‘Association
between regional selenium status and reported outcome of COVID-19 cases
in China’. The American Journal of Clinical Nutrition. 2020;111(6):1297–9.
[70] Kieliszek M. and Lipinski B. ‘Selenium supplementation in the prevention
of coronavirus infections (COVID-19)’. Medical Hypotheses. 2020;143:
109878.
[71] Albert C.M., Hennekens C.H., O’Donnell C.J., et al. ‘Fish consumption and
risk of sudden cardiac death’. JAMA. 1998;279(1):23–8.
[72] Zabetakis I., Lordan R., Norton C. and Tsoupras A. ‘COVID-19: The
inflammation link and the role of nutrition in potential mitigation’.
Nutrients. 2020;12(5):1466.
[73] Iddir M., Brito A., Dingeo G., et al. ‘Strengthening the immune system and
reducing inflammation and oxidative stress through diet and nutrition:
Considerations during the COVID-19 crisis’. Nutrients. 2020;12(6):1562.
Chapter 16
Risk assessment and spread of COVID-19
Challa Sri Gouri1, D. Ajitha2, Nikhil Mulaguru3 and
Goteti Rithika3
Abstract
COVID-19 has brought tremendous changes in everyone’s lifestyle. It also brought
awareness among us on how analysis and prediction of situations play a crucial
role. These kinds of situations and risk assessments are considered critical factors in
reducing the seriousness of the situations. Due to a lack of risk assessment, proper
preventive measures cannot be taken. Generally, if an epidemic occurs throughout
the world and shows its impact on more people, it is declared a pandemic. In a
pandemic situation, the greatest weapon one can use to fight against it is risk
assessment and taking measures. The word risk assessment refers to the procedure
of identifying, evaluating the factors that cause harm to the environment, and living
beings. It also involves making decisions on how to put an end to it. For example, if
we consider COVID-19 pandemic, to manage the risk of spreading, analysis was
done on understanding the situations that increase the risk of transmission of the
virus, identifying the majorly affected people in that situation, coming up with
some solution to stop this from happening or to control the situation. In COVID-19,
the government provided many practical measures such as wearing masks, sani-
tizing our surroundings repeatedly, and maintaining a physical distance. This book
chapter will have clear discussions on the steps taken to assess the risk and stop the
spread of pandemics like COVID-19. We believe that prevention is always better
than cure. The chapter presents an analysis of prevention using prediction techni-
ques. Using sentiment analysis in machine learning, weighted density ensembles,
forecasting models, and risk assessment can be done. Further, a bot named
SAUCHA is proposed for automatic sanitization.
Keywords: Pandemic; Risk assessment; Prediction techniques; Forecasting

models; Weighted density ensembles; Sentiment analysis
1
Department of Electronics and Communication Engineering, Sreenidhi Institute of Science and
Technology, Hyderabad, India
2
School of Computer Science and Engineering (SCOPE), VIT, Vellore, India
3
Department of Mechanical Engineering, CVR College of Engineering, Hyderabad, India
16.1 Introduction
The novel coronavirus disease 2019 (COVID-19) outbreak rapidly spread world-
wide, causing a public health crisis globally [1]. Risk assessment consists of two
words: risk and assessment. Risk can be defined as a situation in which the chance
of people being harmed due to any hazard is high. Assessment refers to the analysis
of factors that can cause harm. The magnitude of the increase in confirmed cases in
an influenza pandemic is a proxy for epidemic size and disease transmissibility [2].
This complete process will help us to identify the problem, understand, and analyze
the reasons and the risks associated with it. Risk analysis includes two major steps.
They are risk evaluation and risk control. Whenever there is an occurrence of any
situation that has a negative impact on the health of the people this risk assessment
is used. This book chapter will discuss the steps taken to assess the risk of COVID-
19 and its spread [3].
Coronavirus belongs to the family of viruses that is capable of causing common
cold to severe health issues. The first case of coronavirus was found in the month of
December. In the month of February, many countries were affected by this making it
an issue of international concern. The name COVID-19 was issued to the coronavirus
disease on the February 11, 2020 by the World Health Organization. It is declared as
a pandemic in the month of March. The word pandemic in medical terminology can
be defined as the condition in which there is a rapid spread of a disease throughout
the world, affecting millions of people. This data shows us how rapid the spread of
the virus was.
Assessing the risk associated with COVID-19 WHO issued few guidelines to
all the countries like maintaining 1-m distance from each other, wearing masks, and
sanitizing regularly. Countries also issued guidelines for lockdown to stop the
transmission and to break the chain of virus. COVID-19, in our opinion, has
become a pandemic, with small chains of transmission in many countries and large
chains resulting in widespread infection in a few countries such as Italy, Iran, South
Korea, and Japan [3]. The findings revealed that controlling infectious disease
transmission is the primary concern in pandemic disease [4]. This chapter gives
information about various methods used for risk assessment and different strategies
for the management of epidemics.
16.2 Technology and epidemics

Pandemics have a massive impact on our lives, economies, and social orders. A
pandemic is characterized as a sickness that quickly spreads in various nations and
countries. Pandemics slaughter vast numbers of individuals. In recent years, we have
seen three lethal pandemics: the 1918 Spanish Flu, 1957 Influenza, and 1968 Hong
Kong Flu [5,6]. The outburst of the 2013 Ebola virus in Africa has affected numerous
lives. Effective regulation of this flare-up depends on fast and viable coordination
and correspondence. Early location and reaction are fundamental to effective control.
However, its utilization in the Ebola outbreak reaction has been restricted. Because of
Risk assessment and spread of COVID-19 423
the requirement for the early disclosure of new EVD cases and immediate treatment,
colleagues passed information to empower a smooth and fast information stream [7].
A successful regulation of external ventricular drain (EVD) depends on a metho-
dology that includes different intercessions: case the board, observation and contact
following, correspondence, and social assembly. In addition, the utilization of active
surveillance, comprising close management and collection of vital signs and critical
clinical indications, is essential to the strong restraint of EVD.
Gene Xpert was initially designed to detect tuberculosis cases but has since
been modified to allow rapid testing of various pathogens, including HIV, malaria,
STIs, and Ebola. It is one of the technologies used to detect Ebola in the DRC. For
example, technicians at the INRB laboratory in Kinshasa can use Gene Xpert to test
for the Zaire strain of Ebola in just 1 h, with the help of USAID, WHO, Canada, the
Global Outbreak Alert and Response Network, and the Emerging and Dangerous
Pathogens Laboratory Network. If a sample is found to be negative, it is tested
again to see if it contains any other Ebola strains, other viral hemorrhagic fevers, or
other diseases [8].
OraQuick is a quick diagnostic test developed in collaboration with the US
Centers for Disease Control and Prevention (CDC) and GOARN. In less than
30 min, OraQuick will test blood or saliva samples for Ebola [9]. Users can enter
the latest data or change the data saved on their phones using the ODK Collect
Application. When the data is finalized, it is submitted. During submission, the
location is automatically integrated with the data. The form hub technology created
forms using the same paper forms that were used to collect data [10]. Whenever
new cases are discovered, the contact tracer enters the information into a new case
on the phone then sends it directly to the form center server. If for any reason, the
phone does not have an Internet connection, the collected data is stored temporarily
on the phones. At the end of the day, when they are returned to the control center,
the phones are connected to the Internet, and stored data is affixed to the server.
The Dashboard technology mainly focuses on providing information based on
proofs to assist the management to make timely decisions. A TV is used to show the
live updates of daily contacts, the interviewers. It can also show symptomatic
contacts as well as maps displaying the location of contacts that have been traced
up in the last 2 days. One of the dashboard segments also shows lab results for
speculated and confirmed cases. The detected contacts and cases within the cities
were plotted with the help of coordinates that were recorded on the ODK forms and
then combined using ArcGIS Mapping software to display the contacts and cases in
the cities. These informed areas concentrated on social mobilization, sensitization,
and awareness generated by social mobilization teams.
COVID-19 is a global pandemic on par with World War II and the
1918 Spanish Flu. For suppressing the COVID-19 pandemic, physical distancing
and self-quarantine are required. To maintain this mandate while preserving the
current state of things, various human behaviors, such as entertainment, education,
meeting, and working, have transitioned from online to offline. This resulted in the
emergence of a variety of digital technologies. There are extensive digital tech-
nologies being used in the COVID-19 epidemic. Based on the different types of
technology, it is categorized into hardware and software. To tackle COVID-19,

roughly 15 hardware and over 50 distinct kinds of software technologies were
deployed [11–13]. Compared with the Spanish flu, SARS CoV-2 has spread very
quickly across every continent in a matter of months. In addition, the use of hard-
ware and software technology remarkably enhanced the healthcare sectors cap-
ability to recognize, track, and contain infected persons.
Hardware technologies, for example, computerized tomography machines in
the medical field and computers, phones, and video-based communication plat-
forms in work, education, and daily life, have brought about an unprecedented
transformation in our lives.
Hardware: The computerized tomography machine is commonly used for
early detection and diagnosis because of the coronavirus’s distinct characteristics.
It is the most widely suggested technology in healthcare sector. The computerized
chest tomography has a higher sensitivity for COVID-19 diagnosis than the RTPCR
test [14]. Using chest computerized tomography devices and deep learning tech-
nology, coronary artery diseases can also be detected and differentiated from
pneumonia and other lung diseases [15]. Various technologies, such as mobile
phones, computers, and robots, are indispensable for remote surveillance and
diagnosis during the epidemic. The majority of coronavirus patients can be mana-
ged remotely, with home quarantine instructions. Though video consultations can
often be conducted over the phone, the video provides visuals and a therapeutic
presence. Mobile devices and webcam attached computer applications like Zoom
and WebEx play a crucial role in providing virtual education, work from home, and
everyday life.
Software: In comparison to hardware technology, many software technologies
are more widely used. Video-based communication platforms are the most
advanced technology in the healthcare domain. COVID-19 patients and those
requiring other normal healthcare services can benefit from phones, tablets, com-
puters, and social media. Video-supported communication platforms, such as
Zoom, WebEx, and Google, meet are digitally safe and effective conferencing in
this particular time. Online video lectures can be carried by using Blackboard,
coursera, and other platforms that enable continuous knowledge distribution for
education. Digital information is used for virtual services data exchange at work
using online surveys, Google Sheets, and other telework tools. Furthermore, social
media platforms such as Twitter, WhatsApp, Facebook, Instagram, and YouTube,
and technologies such as Google Trends and GIS System track, locate, and analyze
outbreaks in real time [16].
Mix use: Aside from hardware and software, there are around five kinds of
integrated technologies used to monitor, prevent, and detect outbreaks, such as IoT,
artificial intelligence, virtual reality, computerized tomography, and the Internet of
Medical Things. The majority of people who use technology are divided into four
categories: healthcare, education, work, and others [17]. There are two types of
users when it comes to technology: providers and receivers. In the healthcare sec-
tor, for example, medical professionals are the suppliers who employ technology to
give service. In contrast, patients and infected individuals are the receivers.
Employers and teachers are examples of providers, whereas employees and stu-
dents are receivers in the education and work sectors. Even though they utilize
diverse types of technology in their daily lives, they all employ the same kind.
16.2.1 Healthcare
During the pandemic, medical professionals and patients with various chronic diseases
are the most frequent consumers of digital technology. On the front lines, radiologists,
surgeons, and nurses are diagnosing and treating patients. The radiologists play a critical
role in diagnosing computerized CT scan for a patient either positive or negative for
COVID-19 disease and describing primary CT characteristics and lesion distribution. In
addition, patients with various chronic conditions, particularly those who have already
been infected with the coronavirus, get services and treatment from healthcare experts
via technology during the same time. Patients with many immune suppressed or chronic
diseases, for example, will have to choose between risking iatrogenic COVID-19
exposure at the time of a doctor visit and delaying required treatments. In addition,
patients must deal with the unavoidable utilization of technology, for example, the
computerized tomography equipment, video communication platforms to acquire
advice and directions from a healthcare expert, whether they choose a direct appoint-
ment, postpone a visit, or use virtual healthcare [18]. As a result, the main groups of
technology users at COVID-19 are healthcare professionals and their patients.
16.2.2 Education
Because of COVID-19, a considerable portion of the population was forced to
study from home to comply with the quarantine obligation. The teachers quickly
adapted to online teaching and put significant effort into preparing for online
classes and engaging the students. The majority of the education sector uses video-
based communication devices and platforms. During this epidemic, students and
teachers have surpassed the general public as the second most frequent consumers
of digital technology [19].
16.2.3 Work
Work professionals have emerged as a significant segment of technology users
during the pandemic. During COVID-19, researchers, scientists, and employees
continue to work remotely using digital technologies. From the work standpoint,
regardless of the technology employed, most work professionals serve as con-
tributors and receivers. An employee, for instance, receives guidance from their
boss while also having to report their work via Zoom.
16.2.4 Others
Apart from the two primary users listed before, health authorities, government
officials, and the public all use technology. To deal with this global spread, indi-
viduals worldwide are becoming more familiar with the technology they use daily.
Public health experts and government officials, for example, employ mobile-based
monitoring technologies to follow the spread of an epidemic or big data
technologies to assess outbreaks and devise policies. At the same time, the general
population receives information via digital devices from all over the world.
16.3 Prediction techniques

After the outbreak of COVID-19, many techniques were used to control its trans-
mission. The control measures taken by various governments include three stages:
screening, suppression, and mitigation. Screening refers to checking the body
temperatures of a person using a thermometer to analyze whether he/she is affected
by the fever caused by the virus. Suppression and mitigation are considered inte-
grated steps that are carried out based on the results of screening. In case he/she has
abnormal temperatures, they can isolate themselves, which helps in avoiding the
transmission of virus. This isolation will also help them in reducing the effect of
viruses on their health that is known as mitigation. Prevention is better than cure. If
we are able to predict the conditions, managing the situation becomes a piece of
cake for us. There are various prediction techniques proposed, which can be
effectively implemented. Various improvements are possible in the existing ana-
lysis, which can help us in the prediction of epidemics.
The three major points we need to consider while discussing the epidemics
caused by animals or other organisms are their toxicity, means of spread, and
measures to be taken. As we are a part of the food chain, toxicity at any level can
have an everlasting impact on us. It can also cause the spread of microbial diseases.
Many microbial diseases or epidemics in the past were also spread due to various
organisms and animals. As per the statistical data, it is very clear that 60% of all
human diseases originate in animals.
Let us have a look at the data indicating reasons for various epidemics:
Rabies—mammals
Anthrax—spore-forming bacteria and infected animals
Arbovirus—arthropod-borne virus
Avian influenza—poultry, animals
B virus—macaque monkeys
Brucellosis—goats, sheep, cow
Cat scratch disease—young cats and kittens
To stop the spread of these kinds of diseases, it is essential to monitor toxicity
at each level. These kinds of diseases caused by animals are known as zoonotic
diseases. These diseases can be airborne, vectors (transmitting infected agents from
animals), close proximity to animals, and foodborne. The motto of discussing these
points in prediction techniques is that when we analyze the past data, we can take
some measures that can help us be ready to face any epidemics in the future.
Few other prediction techniques are as follows:
1. Data analysis
2. Mathematical prediction models
3. Recursive-based prediction model
4. Beesham’s prediction model

5. Boltzmann function-based prediction model
6. Performance evaluation criteria
Data analysis: In prediction techniques, data analysis is considered the major
step. In this step, we collect the data from the past epidemic experiences and start
working on it using various techniques and models. The experts majorly do this
data analysis in that field to give the best results in preventive measures. Working
on data related to epidemics can help us to be prepared for any future situation.
Mathematical model: A mathematical model can be defined as the diagram-
matic representation through which we can analyze the reasons behind a particular
aspect. It has a set of variables and equations that are satisfied by the system.
Recursion-based prediction model: We can analyze the relation between the
number of cases confirmed in a day and the number of cases confirmed on a pre-
vious day. It helps us in relating death and recovery cases [15].
Beesham’s prediction model: Due to the increase in the number of corona
cases, several models were recommended for predicting the COVID-19 outbreak in
the literature using exponential function. In the similar manner, Beesham’s math-
ematical model, which was recommended for considering the confirmed cases of
the COVID-19 in South Africa [20], was also using an exponential function.
A Boltzmann function-based prediction model: The Boltzmann function
helped us by providing a prediction model for the COVID-19 outbreak. The derived
relation, which is represented by an Equation, is related to the sigmoid function.
The exception is linear transform [21].
Another common technique used for prediction is analyzing the condition of the
person affected by the virus. This study helps us in understanding the factors and taking
preventive measures as per the requirement. Wearable devices also play a crucial rule
in monitoring and detecting cases of COVID-19. Devices such as smart watches can be
used to track the health condition of a person. Many of the papers also suggest that
these wearable devices can analyze and predict the virus attack even before seeing the
symptoms analyses modern trends of Internet-based surveillance, taking advantage of
computational science methods: Web queries, Google Trends [22]. We can clearly
understand that various techniques can be implemented to predict epidemics even
before they occur. The major tool for prediction is data. If we successfully gather
enough data from past situations, analysis and prediction are quite simple tasks.
16.4 General methods followed for risk assessment

In the previous sections of this book chapter, we had an elaborate discussion of risk
assessment. In this section, let us discuss various methods used for risk assessment
in detail with their specifications, advantages, and disadvantages.
In general, risk assessment includes the following steps:
1. Identification of the problem
2. Analysis of its effects
3. Assessment of risks
4. Preparing report
5. Reviewing and implementing the measures
While working on risk assessment, we consider the following factors. Risks
generally include health, environment, finance, technology, safety. In general
assessment, we also use 10Ps principle. Risk assessment can be performed in two
ways: individual and group. Individual risk assessment is carried out by considering
each and every individual. This helps us to understand the risk at the root level.
This assessment majorly focuses on the psychological and ideological factors of
individuals. Most of the individuals feel there is no risk for them and their situation
is in control. This type of analysis is considered essential during pandemics. Group
risk assessment is generally performed in a group or large scale where individual’s
conditions are not given much priority. For example, risk assessment during a
cyclone comes under this category. We also name this as system risk assessment.
As we are aware, a system can be linear or nonlinear. When a system is linear,
assessment is easier. In the case of nonlinear as data changes frequently, prob-
abilistic analysis can only be possible.
In organizations, a concept called risk register is used to access the risks. This
register will store all the information related to the risks, which helps that organi-
zation monitor, analyze, and manage the risks. The fisheries science and manage-
ment practitioners at a recent time have begun to carry out formal risk management
processes for fishing activities [23]. A personal perspective and lessons learnt
through risk assessments applied to the management of numerous significant spe-
cies and ecosystems are presented. The significance of engaging and empowering
stakeholders through more intuitive risk assessment methodologies so that they can
actively comprehend and participate in the risk assessment process is one of the key
lessons gained [24]. The development of the “Risk Society,” in which governments,
communities, organizations, and individuals spend much of their day-to-day efforts
managing risk and, where possible, transferring risk to others, has been linked to
the modernization of many nations over the last century, according to prominent
sociologist Ulrich Beck [25].
The health sector’s policies and regulations are becoming increasingly risk-
based. As a result, indicators that focus on the risk to life expectancy, or the average
number of years foregone or added as a result of a proposed policy instrument, are
increasingly routinely used to evaluate health policy [26].
To understand risk assessment in detail, let us discuss the following methods:
1. What-if analysis
2. Fault-tree analysis (FTA)
16.4.1 What-if analysis

As the name suggests, what-if analysis is the process in which a person with the
knowledge on that particular aspect will ask questions to analyze what happens if
some undesired event occurs. To assess the risks associated with a particular
undesirable event that may cause health hazards, one must be aware of its origin,
causes, and consequences and should have complete information about that aspect.
In this analysis technique, people with expertise in that particular topic will sit
together. They start with identifying the problematic situation, its effects, and
methods that can be used to tackle them. They discuss existing solutions, their
disadvantages, and the improvements that are needed. Through this thorough dis-
cussion, they discover opportunities for new methods to be successfully imple-
mented [27]. The steps that are followed by them are as follows:
1. Presenting detailed diagrams and guidelines
2. Prepare questions
3. Analysis and assessment of risk
4. Discussion
5. Summary of the major points
6. Preparation of plan of action
Limitations of this method:
1. Majorly depends on the questions that are framed.
2. Active participation of members is essential.
3. This method is considered introspective.
4. Reviewer plays a major role.
Advantages of this method:
1. Simple and easy
2. No requirement of specialized tools
16.4.2 Fault-tree analysis

This is a method in which we understand the reason for a failure due to which we
get undesired outcomes or events. In FTA, Boolean logic is used. This system is
mainly used in various engineering streams to detect the reasons behind systems
failure and decrement the chances of risks associated with that. Aerospace engi-
neers majorly employ this technique during designing [28]. FTA technique was
developed in the year 1962 in Bell laboratories. At present, it is used in reliability
systems for risk assessment. In this analysis, by creating a logic diagram, we will be
able to figure out the relation between fault, subsystem, and safety design elements.
The system’s failure is considered the root of the tree, which represents the
outcome that is undesired. While we are working on top-down analysis, this failure
may happen due to two reasons. They are mistakes during normal operation or errors
in the maintenance operation. Any error among these two will result in failure. So it
is considered OR logic. Again considering the normal operation, it has two branches,
which leads to another OR logic. To overcome the failure, improvements must be
made in the design that is considered logical AND. By programming these logics, we
can get the probabilities of failure using the FTA technique.
Early on, we have seen human errors and negligence causing disasters [29].
There are few algorithms and analysis that help in predicting the probability of
errors occurring. One such analysis is the FTA [30]. FTA is an analysis tool that
helps identify faulty events and would result in damages and disasters if not cor-
rected on time [31]. It can be said that FTA is a hierarchical model defining the
topmost columns—the ones that are the most important operations, which include
various fields of production, distribution, maintenance based on the implementation
area [32]. Thus, making the FTA defined events as a set of true or false assertions,
cutting down all the undesirable conditions using the logic gates [33].
It would be a good practice to implement FTA before initiating any projects or
activities to overcome hurdles later on and can avoid great tragedies. Some of the
applications of FTA are reliable power systems [34]. Gas leaks from the industrial sites
can be halted [35]. Usage of hazardous methods in the production can be seized in its
initial stage [36]. Implementations of both Environmental Impact Assessment (EIA) and
FTA together bring awareness in the public about the environment and help in detecting
the areas of errors and prevent it from happening, ensuring a secure environment [8].
To represent these logics in FTA symbols like basic event, external event,
undeveloped event, conditioning event, intermediate event, let us have a brief
discussion on the steps included in FTA:
1. Identification of undesired event
2. Understanding the working of system
3. Designing the fault tree
4. Analyzing the fault tree
5. Assessing the risks
Advantages of FTA:
1. It is suitable for computer programs.
2. Capable of finding faults in different paths.
3. Effective in analyzing the reasons for faults.
Let us also discuss the guidelines that are issued by the World Health
Organization to control this situation.
16.4.3 Guidelines issued by World Health Organization

WHO issued the guidelines based on the phase in which they are. The phases are as
follows:
1. Providing information to the people and bringing awareness to them.
2. Developing proper understanding of people, which makes them ready to face
any situation.
3. Taking proper measures so that people can adapt themselves to the situation.
4. Measures to be taken and steps to be followed.
5. How to handle activities after and during the epidemic.
Hand hygiene, respiratory hygiene, and basic food safety procedures are
among the WHO’s basic recommendations to the general population to avoid
exposure to and transmitting a variety of diseases [37]. Furthermore, providing
supportive care to infected individuals can be pretty helpful [38].
The chance of contracting COVID-19 is determined mainly by age, gender, and

other conditions (obesity, diabetes, lung disease, kidney disease, and heart disease) [39].
The key points of the mentioned guidelines are as follows:
1. Avoiding social gatherings
2. Maintaining social distance
3. Wearing masks
4. Sanitizing regularly and repeatedly
5. Providing accurate and regular updates to people.
16.5 Prevention and management of epidemics

Prevention can be defined as the steps that we take to stop or prevent something
from happening. Management with respect to epidemics can be defined as the
measures that we take to control the spread and transmission of disease. In the
epidemics, subject prevention and management are considered key factors.
Prevention is the step that we take to avoid the disease from occurring.
Management is the step that we take to control the effect or impact of the epidemics
after it occurs. In this section, we will have an elaborate discussion on various
preventive measures and management techniques. The strategies proposed in this
regard are focused on one of the best analysis techniques of machine learning, that
is, sentimental analysis.
During the ferocious outburst, clinical specialists have been attempting to
develop novel treatments and vaccines, and scientists were working in the fields of
data science and technology have been attempting to discover the infectious agent
and aid in its control through the use of information-based methods [40].
Despite advances in medical research in treatment, pathogen discovery, and
diagnostics, infections can still spread widely and cause serious health con-
sequences. Modern technology, on the other hand, has introduced new approaches
and tools [41].
The general preventive measures that can be taken to protect ourselves from
COVID-19 are as follows:
1. Do not touch your face.
2. Sanitize your hands regularly and repeatedly.
3. Maintain distance from others. Do not shake hands.
4. Use a napkin while sneezing or coughing to cover your mouth and nose.
5. Never use other’s belongings.
6. Avoid going to crowdie places and social gatherings.
7. Wash everything before using.
8. Wear a mask.
9. Self-isolation can be considered the best measure in case if you feel you
are sick.
Management of epidemics: If we can predict the epidemics through the techni-
ques that we discussed in the previous section, like what-if analysis and fault-tree
analysis, management becomes very easy as we can understand and analyze the
situation very easily. Management completely concentrates on stopping the spread of
virus and reducing the number of people affected by it. Epidemic preparedness is very
crucial in this regard. Another crucial factor to be considered is the management of
epidemics that comprises combination of various sectors like public and health. Active
involvement and response from all the sectors are also important. Coordination and
cooperation between various sectors is crucial in effective management.
The causes of epidemics may be different but most of the effects are almost
similar. Even during this COVID-19, one of the countries was able to effectively
manage the situation. As a result, the rate of transmission and number of corona
deaths were very low when compared to all other countries. The reason behind this
is that particular already faced this kind of situation which helped them in pre-
dicting and taking measures as per the requirement.
Various strategies are employed for the effective management of epidemics.
Let us explore them.
16.5.1 Strategies proposed

When compared to all other countries, Africa is the country that is least affected by
COVID-19. The main reason behind this is the steps taken for the early detection of
cases and self-isolation. When we are discussing the strategies for effective man-
agement following points must be considered:
1. Health and public sector workers must be provided support and their safety
must be considered a major priority.
2. Proper maintenance of hospitals.
3. Developing strategies that will help us in making our health system strong.
4. Strategies must be framed considering individual case studies.
5. Active response and communication between various departments for effective
management of the situation.
6. Food, water, and health facilities must be properly provided.
Strategies of various countries
Japan: In Japan, they have followed the strategy of identifying the con-
taminant zones. In these zones, they doubled the rate of tests. Strict rules were
made to stop the travelers from other places to these contaminant zones. The major
point is the cost of treatment that is completely covered by the government [42].
South Korea: They employed a screening strategy in huge number for the
individuals with symptoms. They were successful in bringing awareness to people
about thermal screening and sanitization. Strict rules were imposed and social
gatherings were banned.
By considering various strategies implemented by different countries effective
strategies are listed as follows:
1. Maintaining physical distance
2. Tracing the infected person and his contacts
3. Performing corona tests
4. Improvement facilities and staff in hospitals

5. Pre-supply of PPE kits
6. Proper communication and coordination
7. Immediate response
16.5.2 Sentimental analysis using machine learning

The term sentimental analysis refers to a tool used to analyze the emotions in text.
Using these machines can analyze feelings and emotions without human inter-
ference. The models are trained to understand the feelings hidden behind the text,
such as anger and sarcasm.
The prominence of rapidly growing online social networks and media-based
societies has impacted young researchers to work on sentiment analysis. These
days, most businesses depend on customers or people’s opinions regarding their
products from social media [43]. Due to the lack of opinion text present in the
digital form, there was very less working interest on computational linguistics in
the end decade of the twentieth century was noticed [44].
Working of this technique: Various techniques and algorithms are used to
train the models to perform sentimental analysis. The Bayes theorem of probability
can be used to check whether a word is given in positive or negative context:

A B P ðA Þ
P ¼P
B A P ðB Þ
In this, words can generally be calculated against each other. Using the trained
models, we can analyze whether the word is positive or negative.
Apart from the Bayes theorem of probability, the other algorithms used in the
analysis are as follows:
1. linear regression,
2. support vector machines.
Deep learning can also be used when multiple algorithms are used at the
same time.
For example, if you type “I am so sad,” using the sentimental analysis datasets,
the machine will understand that sad is a word that describes negative emotion.
● Use of sentimental analysis in management of COVID-19
In the present situation, social media is playing a vital role. This sentimental
analysis can be performed on any social media handles like Twitter, Instagram, and
Facebook [45]. For this analysis, we require two stages as follows:
1. Dataset preparation stage
2. Analysis stage
Using this, we can analyze the intensity of words and the polarity of words,
which helps to know the condition of a person. This method can help us to reduce
his/her fear and make them mentally strong when they are affected by COVID-19.
This method will help us in fighting effectively with the coronavirus. Using this
analysis, public statement can be analyzed and we can take measures according
to that to strengthen them. In this analysis process, first social media platforms
are tracked. Then data is collected through them. Once the collection of data is
processed, data analysis is performed to analyze the sentiments behind the
statements.
16.6 Protecting the living beings from the impact of

epidemics
The epidemics have different kinds of impacts on various fields. In the case of
COVID-19 pandemic, it affected the lives of teachers, students, migration workers
and laborers. It has a negative impact on each and every field. It has affected the
economy very badly. Apart from these things, people’s mental health is also getting
disturbed as they are restricted to their houses. In terms of economy every family is
facing problems as they are unable to get their salaries on time, prices of essential
commodities and vegetables increased suddenly. Most of the people are suffering
from food scarcity.
The impacts of COVID-19 are much more when compared to the two world
wars. In the two world wars only few countries participated but almost all the
countries were affected due to this COVID-19. In Italy and the United States, the
numbers of deaths due to coronavirus were very high. Let us discuss the impacts on
each field in a detailed manner.
16.6.1 Impact of COVID-19 on agriculture sector

This pandemic has had a devastating effect on the agriculture sector. It affected the
production and marketing of food items. This also affected the prices of com-
modities. The increase in the prices of commodities has a severe impact on the
consumption of food. We also know that many of the states were affected by locust
attacks during this crisis period, which can also be considered one of the major
reasons affecting the production rate. Even though relaxations were given to the
agriculture sector due to labor scarcity, good harvest was not possible. Due to
restrictions imposed on transportation and marketing of products during the lock-
down, many food products were damaged as per the government reports [35].
16.6.2 Impact of COVID-19 on economy

COVID-19 majorly impacted the economy of several countries. Pandemic showed
its impact on GDP, stock market, oil prices, global expenses, etc. It was wreaking
havoc on financial markets. It was a threat and can be compared to the phase of the
Great Depression that occurred after the two world wars. Many countries suffered
in different ways. If we consider Japan, its main stock index saw the lowest mark
since 2016. In the case of Spain, the main stock exchange drastically dropped. It
was 10,000 points before. Due to the pandemic situation, it is only under 7,000
points. Boeing lost 18% of its value that can be considered very high [46].
16.6.3 Impact of COVID-19 on educational sector

The most severely affected is the educational sector. Due to the pandemic, the
system of education is entirely changed. It is shifted to complete online mode. This
has its own pros and cons. Majorly, the children in villages with issues related to
the Internet, PCs, and laptops lose interest in classes, resulting in an increase in
dropouts. Most of the schools are not taking these online classes too. This way of
learning completely depends on the dedication and self-motivation of the students.
Promoting the students to next classes without exams would completely affect the
standards of education they are pursuing. This will cause a lack of seriousness
among students [47]. Recovery from these impacts will take much time. It needs
many efforts. This pandemic leads to the digitalization of the world. Even post-
pandemic, there will be many changes in the existing systems. We need to adapt
ourselves to these situations and accept the changes. We must implement the
leanings of these epidemics and analyze the situation to tackle future situations
effectively. We must not neglect them after the pandemic. We must work on them.
Technologies must be used effectively to come up with the best solutions.
16.7 Our contribution
The major problem to everyone after the COVID-19 is fear of bacteria, germs, and
viruses. The extreme places for germs, bacteria, and virus are through the floor.
Even though in the major places like hospitals and malls they clean regularly, there
will be germs and bacteria on the floor itself as there will be many different people
visiting there. Some of them become airborne and spread. So, when they touch our
feet or shoes and when we kick up that can spread to the air we breathe and can
cause illness to us.
16.7.1 Proposed method and its working

Health is considered wealth, and to ensure safe health, we have come up with a bot
that detects germs, bacteria, and viruses and sanitizes a person before entering a place
like a hospital or shopping mall. The working of the bot is when a person stands on it,
his legs/shoes will be completely sanitized by ultraviolet rays and shown on the LCD
placed on the bot. Once the feet/shoes are completely sanitized using ultraviolet rays
directly, a person will get sanitizer to sanitize his hands perfectly. We are attaching
an automatic stand for our bot, which can be closed or closed opened based on the
algorithm. So, the person’s hands and legs will be virus-free and ensure no germs
spread through floors. Our bot also has an add-on feature that is it can even sanitize
the floor of hospitals as we are attaching freely movable tyres to it. These tyres can
even be closed once they are done with cleaning the floor. The functionality of
SAUCHA is clearly indicated in Figure 16.1.
The major problem to everyone after the COVID-19 is fear of bacteria, germs,
and viruses. The extreme place for the germs, bacteria, and virus are through floor.
Even though in the major places like hospitals, malls they clean regularly there will
It can sanitize the Onestep for cleaning

surroundings in an SAUCHA-sanitizing germs, viruses, bacteria
effective manner with BOT present on the footwear
easy operation before entering
Once successfully done, Provides sanitizer for

the person gets indication cleaning hands and other Once it is done. Gives
to enter accessories if required alert to the user
Figure 16.1 Functionality of SAUCHA
be germs and bacteria on the floor itself as there will be many different people
visiting there.
UVC radiation is a known disinfectant for air, water, and nonporous surfaces.
This radiation has effectively been used for decades to reduce the spread of bac-
teria, such as tuberculosis. For this reason, UVC lamps are often called “germici-
dal” lamps. This UVC LED plays a vital role in killing the germs and bacteria on
the surfaces. Nowadays, people show a great interest toward technology and every
human-made things can be achieved through technology. So a robot controlled
through Bluetooth will have a great impact on attracting the people as well as its
very convenient to the user.
As we know that even after COVID-19 sanitization plays a vital role, most of
the people think about it, so this UVC sanitization had an effective sanitization and
helps in killing germs and bacteria, and an autonomous hand sanitizer is also very
important to avoid common contact, which helps in preventing the spread of
the virus.
16.7.2 Components required

1. Arduino Uno
2. UV-LED
3. Ultrasonic sensor (hcsr04)
4. LCD display (162)
5. IR sensor
6. Servo motor
7. Bluetooth module
8. Relay module
9. 12-V batteries
10. Wheels
11. Dummy motors
12. Side shaft motors
13. L298n motor driver
14. On/off switch
15. L-clamps
16. Jumper wires
These components are connected together for the implementation of SAUCHA
as shown in Figure 16.2.
16.7.3 Software required and simulation

1. Arduino IDE
2. Proteus
To clearly observe the structure of bot, the CAD model of the proposed method
is shown in Figure 16.3.
Arduino Ultrasonic sensor L298n motor driver
Bluetooth module IR sensor UV-LED, LCD
Figure 16.2 Block diagram of SAUCHA
Figure 16.3 3D model of SAUCHA

Figure 16.4 Interfacing of components
To verify the simulation results in the bot, the interfacing of the components in
Proteus software is connected as shown in Figure 16.4.
The bot LCD is connected to Arduino. LCD is used for displaying the
instructions and information. An ultrasonic sensor is used to measure the distance
between the person and the bot. UV-LED is used to disinfect the person. A motor
driver is used to control the motors and the wheels are used for the movement of the
bot. The servomotor is used for flipping UVC LEDs, and a servomotor is also used
in a sanitizer dispenser. An IR sensor is used for obstacle detection during the
movement of our bot. Relay is used for controlling the circuit. Arduino IDE is used
for writing and uploading the code in to the board. Proteus is the software used for
the simulation of our circuit. When a person is detected, the ultrasonic sensor
detects and drives the motor to rotate the sanitizer dispenser that will open and it
will sanitize the person. Bluetooth module that is connected to Arduino can be used
to transfer data which can be an add-on feature for future scope to access the data
from bot.

This pandemic as discussed brought many changes in various fields. With the
occurrence of pandemics, people start realizing the importance of health, hygiene,
and well-being. The bot proposed in this chapter can be installed in various places
where more people are moving and interacting to sanitize their hands and foot to
avoid the spread of COVID. In future the per capita income on health sector would
increase. Apart from that the individual expenditure on products related to health
sector also increases. The sector that is least affected by the COVID-19 is health
sector. Apart from that tremendous changes occur in different sectors. Mainly
educational sector will witness many changes that will majorly concentrate on
development of skills. Pandemic helped many innovators to come up with their

wonderful innovations. In this year itself under the MSME, many start-ups were
registered. Many industries used this crisis time as an opportunity to analyze
themselves and bring changes as per the requirement. This also helps different
countries analyze their mistakes and work on prediction, epidemic management
techniques, and the improvements that need to be done in the health sector.
Preparation and proper planning can be implemented to have preparedness to face
any epidemics in future. There is also scope for new technologies in epidemic
management.
Epidemics have much impact on society when compared with the world wars
that occurred in the past. The major key to successfully facing these epidemics is
exact data about that particular disease, effective management strategies, and
improvement in the field of technology that can help us work much on prediction
and prevention. The major reason behind failure in effectively managing the
situation during epidemics is lack of proper information, neglecting the measures to
be taken, no proper awareness among people.
Various machine learning techniques provide us help in analyzing the
situation, mental health of people. During times of epidemics, the government
must make sure that fake news is not spread on social media. Even it should
restrict the news channels. The media must provide exact information to people,
but it should not panic them. When we analyze the death reports of people, we
understand that the major reason was their fear and psychological conditions
rather than the effect of the virus. During such times, situations must be dealt
with sensitively. Researchers have been developing signaling for very low power
consumption [48]. Furthermore, introducing the IoT models allows identifying
flagged, probable, or proven infected cases or a flagged area or object [49], to
take precautionary steps. COVID-19 or 2019-nCoV is no longer pandemic, but
rather endemic, with more than 651,247 people around world who have lost their
lives after contracting the disease [50]. On the economic front, the major fore-
seen impact of AR/VR in intraoperative navigation would be reducing time of
operative procedures, infection rates, revision rates, and length of stay [51].
Contact tracing was one of the important strategies used for reducing, halting,
and reversing deaths and infections caused due to COVID-19 [52].
References
[1] Wu Z and McGoogan JM. Characteristics of and important lessons from the
corona virus disease 2019 (COVID-19) outbreak in China: Summary of a
report of 72314 cases from the Chinese center for disease control and pre-
vention. JAMA. 2020; 323: 1239–42 10.1001/jama.2020.2648.
[2] Chong KC, Zee BCY and Wang MH. Approximate Bayesian algorithm to
estimate the basic reproduction number in an influenza pandemic using
arrival times of imported cases. Travel Medicine and Infectious Disease.
2018; 23: 80–6.
[3] WHO Coronavirus disease 2019 (COVID-19) situation report—44.

Pandemic Influenza Risk Management Guidance, 2020. https://www.who.
int/docs/default-source/coronaviruse/situation-reports/20200304-sitrep-44-
covid-19.pdf?sfvrsn¼783b4c9d_2.
[4] Adhikari SP, Meng S, Wu Y-J, et al. Epidemiology, causes, clinical mani-
festation and diagnosis, prevention and control of coronavirus disease
(COVID-19) during the early outbreak period: A scoping review. Infectious
Diseases of Poverty. 2020; 9(1): 29.
[5] Qiu W, Rutherford S, Mao A and Chu C. The pandemic and its impacts.
Health, Culture and Society. 2017; 9: 1–11. 10.5195/HCS.2017.221.
[6] Ahn I, Jeong B-J, Bae S-E, Jung J and Son HS. Genomic analysis of influ-
enza A viruses, including avian flu (H5N1) strains. European Journal of
Epidemiology. 2006; 21: 511–9.
[7] Bempong NE, De Castañeda RR, Schütte S, et al. Precision global health –
The case of Ebola: A scoping review. Journal of Global Health. 2019; 9(1).
[8] World Health Organization. New technology allows for rapid diagnosis of
Ebola in the Democratic Republic of the Congo. 2017.
[9] Ahn I, Jeong BJ, Bae SE, Jung J and Son HS. Genomic analysis of influenza
A viruses, including avian flu (H5N1) strains. European journal of epide-
miology. 2006; 21(7): 511–9.
[10] Tom-Aba D, Olaleye A, Olayinka AT, et al. Innovative technological
approach to Ebola virus disease outbreak response in Nigeria using the open
data kit and form hub technology. PLoS One. 2015; 10(6): e0131000.
[11] Bhuyan HK, Chakraborty C, Shelke Y and Pani SK. COVID-19 diagnosis
system by deep learning approaches. Expert Systems. 2021: e12776. https://
doi.org/10.1111/exsy.12776.
[12] Dash S, Chakraborty C, Giri SK, Pani SK and Frnda J. BIFM: Big-data
driven intelligent forecasting model for COVID-19. IEEE Access. 2021: 1–
13. 10.1109/ACCESS.2021.3094658.
[13] Aslam B, Javed AR, Chakraborty C, Nebhen J, Raqib S and Rizwan M.
Blockchain and ANFIS empowered IoMT application for privacy preserved
contact tracing in COVID-19 pandemic. Personal and Ubiquitous
Computing. 2021: 1–17. 10.1007/s00779-021-01596-3.
[14] Geldsetzer P. Use of rapid online surveys to assess people’s perceptions
during infectious disease outbreaks: A cross-sectional survey on COVID-19.
Journal of Medical Internet Research. 2020; 22(4): e18790.
[15] Fanelli D and Piazza F. Analysis and forecast of COVID-19 spreading in China,
Italy and France. Chaos, Solitons and Fractals. 2020; 134. article 109761.
[16] Vargo D, Zhu L, Benwell B and Yan Z. Digital technology use during
COVID-19 pandemic: A rapid review. Human Behavior and Emerging
Technologies. 2021; 3(1): 13–24.
[17] Bempong N-E, De Castañeda RR, Schütte S, et al. Precision Global Health –
The case of Ebola: A scoping review. Journal of Global Health. 2019; 9(1).
[18] Greenhalgh T, Koh GC and Car J. Covid-19: A remote assessment in pri-
mary care. BMJ. 2020; 368.
[19] Dhawan S. Online learning: A panacea in the time of COVID-19 crisis.

Journal of Educational Technology Systems. 2020; 49: 5–22.
[20] Beesham A. Mathematical modelling of COVID-19 in South Africa. Asian
Pacific Journal of Tropical Medicine. 2020; 13(5): 235.
[21] Fu X, Ying Q, Zeng T, Long T and Wang Y. Simulating and forecasting the
cumulative confirmed cases of SARS-CoV-2 in China by Boltzmann function-
based regression analyses. Journal of Infection. 2020; 80(5): 578–606.
[22] Google. Google Trends. 2018. https://trends.google.com/trend (Accessed 11
Nov 2018).
[23] Hobday AJ, Smith ADM, Stobutzki IC, et al. Ecological risk assessment for
the effects of fishing. Fisheries Research. 2011; 108(2–3): 372–84.
[24] Fletcher WJ. Review and refinement of an existing qualitative risk assess-
ment method for application within an ecosystem-based management fra-
mework. ICES Journal of Marine Science. 2015; 72: 1043–56.
[25] Beck U. Risk Society: Towards a New Modernity, 1992, Sage, New Delhi.
[26] Yokota F and Thompson KM. Value of information analysis in environ-
mental health risk management decisions: Past, present, and future. Risk
Analysis. 2004; 24: 635–50.
[27] Golfarelli M, Rizzi S and Proli A. Designing what-if analysis: Towards a
methodology. DOLAP: Proceedings of the ACM International Workshop on
Data Warehousing and OLAP. 51–58. 2006. 10.1145/1183512.1183523.
[28] Patil R. An overview of fault tree analysis (FTA) method for reliability
analysis. Journal of Engineering Research and Studies. 2013; 4: 6–8.
[29] Aini MS and Fakhru’l-Razi A. Latent errors of socio-technical disasters: A
Malaysian case study. Safety Science. 2013; 51(1): 284–92.
[30] Shalev DM and Tiran J. Condition-based fault tree analysis (CBFTA): A new
method for improved fault tree analysis (FTA), reliability and safety calcula-
tions. Reliability Engineering & System Safety, 2007; 92(9): 1231–41.
[31] Ericson II CA. Fault tree analysis. System Safety Conference, Orlando,
Florida. vol. 1. 1999.
[32] Khan FI and Haddara MR. Risk-based maintenance of ethylene oxide pro-
duction facilities. Journal of Hazardous Materials. 2004; 108(3): 147–59.
[33] Guth MAS. A probabilistic foundation for vagueness and imprecision in
fault-tree analysis. IEEE Transactions on Reliability. 1991; 40(5): 563–71.
[34] Volkanovski A, Čepin M and Mavko B. Application of the fault tree analysis
for assessment of power system reliability. Reliability Engineering & System
Safety. 2009; 94(6): 1116–27.
[35] Cariappa AGA, Acharya KK and Adhav CA. Impact of COVID-19 on the
Indian agricultural system: A 10-point strategy for post-pandemic recovery.
Outlook on Agriculture. 2021. https://doi.org/10.1177/0030727021989060.
[36] Hamka MA. Safety risks assessment on container terminal using hazard
identification and risk assessment and fault tree analysis methods. Procedia
Engineering. 2017; 194: 307–14.
[37] World Health Organization, “Q and A on coronaviruses.” https://www.who.
int/fr/news-room/q-a-detail/q-acoronaviruses.
[38] D’Antiga L. Coronaviruses and immunosuppressed patients: The facts dur-

ing the third epidemic. Liver Transplantation. 2020; 26(6): 832–4.
[39] Fanelli D and Piazza F. Analysis and forecast of COVID-19 spreading in China,
Italy and France. Chaos, Solitons and Fractals. 2020; 134. article 109761.
[40] Atif I, Cawood FT and Mahboob MA. The role of digital technologies that could
be applied for prescreening in the mining industry during the COVID-19 pan-
demic. Transactions of the Indian National Academy of Engineering. 2020: 1–12.
[41] Johnson HA, Wagner MM, Hogan WR, et al. Analysis of web access logs
for surveillance of influenza. Medinfo. 2004; 11(Pt 2): 1202–26.
[42] Sayeed UB and Hossain A. How Japan managed to curb the pandemic early
on: Lessons learned from the first eight months of COVID-19. Journal of
Global Health. 2020; 10: 020390. PMCID: PMC7688188 PMID: 33282218.
[43] Parvathy G and Bindhu JS. A probabilistic generative model for mining
cybercriminal network from online social media: A review. International
Journal of Computers and Applications. 2016; 134(14): 1–4. doi:10.5120/
ijca2016908121.
[44] Cambria E and White B. Jumping NLP curves: A review of natural language
processing research. IEEE Computational Intelligence Magazine. 2014; 9
(2): 48–57. doi:10.1109/mci.2014.2307227.
[45] Alamoodi AH, Zaidan BB, Zaidan AA, et al. Sentiment analysis and its
applications in fighting COVID-19 and infectious diseases: A systematic
review. Expert Systems with Applications. 2021; 167: 114155. PMCID:
PMC759187 PMID: 33139966.
[46] Verma P, Dumka A, Bhardwaj A, Ashok A, Kestwal MC and Kumar P. A
statistical analysis of impact of COVID19 on the global economy and stock
index returns. SN Computer Science. 2021; 2.
[47] Pokhrel S and Chhetri R. A literature review on impact of COVID-19 pan-
demic on teaching and learning. Higher Education for the Future. 2021; 8.
https://doi.org/10.1177/2347631120983481.
[48] Chakraborty C, Gupta B and Ghosh SK. A review on telemedicine-based
WBAN framework for patient monitoring. Telemedicine Journal and e-
Health. 2013; 19(8): 619–26. ISSN: 1530-5627, 10.1089/tmj.2012.0215.
[49] Garg L, Chukwu E, Nasser N, Chakraborty C and Garg G. Anonymity pre-
serving IoT-based COVID-19 and other infectious disease contact tracing
model. IEEE Access. 2020; 8: 159402–14. 10.1109/ACCESS.2020.3020513,
ISSN: 2169-3536.
Mohammed IA. Supervised machine learning models for prediction of COVID-
19 infection using epidemiology dataset. SN Computer Science. 2021; 2(11): 1–13.
[51] Shelke Y and Chakraborty C. Augmented reality and virtual reality trans-
forming spinal imaging landscape: A feasibility study. IEEE Computer
Graphics and Applications. 2021; 41(3): 124–38. 10.1109/MCG.2020.300035.
[52] Roy S and Chakraborty C. Panic buying situation during COVID-19 global
pandemic. Journal of Information Technology Management. 2021; 13(2):
231–44. doi:10.22059/jitm.2021.80625.
Index
ACE2: see angiotensin-converting model, feature of 380

enzyme 2 (ACE2) MQTT 365
ACGAN: see auxiliary classifier NodeMCU 369
generative adversarial networks open-source platform 360–1
(ACGAN) pandemic and post pandemic
acute respiratory distress syndrome situation 381
(ARDS) 408, 411, 413–14 proposed robot
adaptive neuro-fuzzy inference system bot, function of 376
(ANFIS) 37–8 ultrasonic observations 377
AlexNet 272–3, 275 value proposition 377–8
ANFIS: see adaptive neuro-fuzzy prototype 366, 379–80
inference system (ANFIS) ResNet 367
angiotensin-converting enzyme 2 RFID 366
(ACE2) 297, 299–301, 323, robot model 379
407 social distancing 368
ANN: see artificial neural network spectrum, band of 370–1
(ANN) ultrasonic sensors, interfacing of
ARDS: see acute respiratory distress 377–8
syndrome (ARDS) using AI/ML techniques 380
Arduino-based far UVC 381 UV blaster 371
alpha, features of 362 WBAN 367
analogue pins 361–2 Arduino Uno 373–4
Bag valve masks 370 ARIMA model: see autoregression
corona cleaner-source 371 integrated moving average
COVID-19 pandemic, outbreak of (ARIMA) model
377 artificial neural network (ANN) 37,
disinfection process 368 71, 367
far-UVC lamps 360, 363–4 ascorbic acid (vitamin C) 413
hardware and software designs 361 autoregression integrated moving
improvements and requirements average (ARIMA) model 147
hardware requirements 372–5 auxiliary classifier generative
software requirements 375 adversarial networks
IoT system 365 (ACGAN) 276
IR sensors 370
LIDAR 379 BDAs: see big data analytics (BDAs)
MDH 365 bee products 412
Beesham’s prediction model 427 microarray and RNA-seq data

big data analytics (BDAs) 34, 51 230–1
big data implementation, process of modalities of, medical imaging 231
40–1 biomedical waste (BMW) 386, 388–9
computational approaches 40 disinfection and disposal techniques
COVID-19 outbreak, control and for 386, 391–2
monitoring of 42 healthcare waste 392
drugs manufacturing models, blockchain technology 20–2
production of 44 applications 144
environmental regions 42 categories 143
evidence-based actions 40 challenges of 18–19
guiding principle, techniques of 43 chronological data transactions,
integrated big data conceptual record of 142
model 41 components
outbreak reports 42–3 applications of 6–7
using, limitations of 44–8 block 6
vaccine, development of 43 consensus mechanism 8
vaccine research 43–4 difficulty target 8
biomedical imaging 254–5 header 6
application of 228–9 Merkle root 8
array data analysis 228 nonce 8
biomedical applications 230 previous hash 8
biomedical data analysis 228 structure 7–8
ChiP-seq data 231 timestamp 7
computational algorithms, version 7–8
categories of confront COVID-19
array-based data analyses 233–5 contact tracing 15–16
computer-assisted diagnosis database security 16
technique 232 data fabrication, prevention of 17
hybrid data analysis 235, 250–1 information sharing 16–17
COVID-19 pandemic, role CT in IoMT 18
238, 249, 251–2 COVID-19 healthcare
COVID-19, smart technologies in blockchain-network Electronic
building ML models 253 Medical Record system 161–2
healthcare, IoT impact in 253–4 healthcare data problems 163
de novo mutation 231, 249 HealthyBlock 162
diagnosis using, different techniques COVID-19 pandemic data, real-time
for analysis of 165–6
brain 235–6 African countries, COVID-19
breast 236 tests and cases in 168–9
kidney 236–7 COVID-19 cases and vaccination
ovary 237 status 166–8
skin cancer 237–8 SIR model 163–4
soft tissue sarcoma 238, 247–8 standard logistic regression
DNA microarray 229 model 164
Index 445
time-to-event analytics model celastrol 411

164–5 chest X-ray (CXR) 118, 263, 367, 369
total cases vs. death and recovery CI: see computational intelligence (CI)
records 168–70 circular economy (CE) 87, 89, 100–3
data handling 172–3 CLI: see command line interface (CLI)
healthcare data records, better cloud-based IoMT 81–2
exchange of 152 architecture of, proposed IoT
healthcare, Internet of Things framework 73
(IoT) in 153–4 cloud infrastructure connects
interoperability 155 74–6
medical research and supplies, data analysis center 74
validating trust in 152–3 physicians 74
optimized privacy and data quarantine and isolation, center
security 154 for 74
potential benefits of 151–2 symptom data, collection and
privacy 156–7 upload of 74
prospects of 157–60 collecting real-time symptom data
scalability 156 using machine learning
security 155 algorithms
speed 155 data preprocessing 70–1
stringent data protection dataset 70
regulation 155–6 predictive model 71–2
validating correct billing data acquisition using wearables
management 153 devices 76–7
distributed ledger 145–6 Google Assistant response
global market size of 146 cough symptoms 79
healthcare administration 149–50 cumulative reports 81–2
healthcare applications 171 high fever symptoms 79, 81
healthcare data exchanges 149 no symptoms 79–80
healthcare data systems 147–8 person having fever 79–80
healthcare data using, management implementation of, proposed
of 144–5 framework
medical information management IFTTT web service 76, 78
143 using Google Assistant and
overview 6 Adafruit API 76, 78
pharmaceuticals 150–1 cloud service 199
possible solutions and deliverables CLR: see critical literature review
19–20 (CLR)
types of 8–9 CML: see continual machine learning
BMW: see biomedical waste (BMW) (CML)
Boltzmann function-based prediction CNNs: see convolutional neural
model 427 networks (CNNs)
cognitive rehabilitation, stress
call center 196, 198 detection for 351
CE: see circular economy (CE) data acquisition 334
dataset preparation 344–5 supervised learning 38

discrete wavelet transform–based types of 35
feature extraction outcome unsupervised learning 38–9
345–7 using, limitations of
EEG-based antidepressant 334 CI-based radiological innovation
mental stress 332 45
performance evaluation 332, CORD-19 dataset 46
348–50 COVID-19 epidemic, issues in 45
performance metrics 348 information-sharing platforms
principal component analysis–based 47–8
dimensionality reduction risks 47
outcome 346–7 visitor information, importance
proposed framework 335, 340 of 46
classification using support vector computerized tomography (CT) scan
machine 343–4 43, 45, 116, 119, 128, 262
EEG 340 continual machine learning (CML)
feature extraction using DWT 137
341 continuous wavelet transform (CWT)
PCA, working mechanism of 341
342–3 convolutional neural networks
three-layer wavelet (CNNs) 128, 130, 264–6,
decomposition 342 270–3, 367, 369
RBF 335 COVID-19 290–1
sLORETA-based activated brain available vaccine
region selection 345 subunit approach 66
state-of-the-art literature 335–9 viral vector vaccine 65
statistical significance using t-test whole-microbe approach 65
350–1 biomedical imaging: see biomedical
support vector machine–based imaging
classification outcome 346, cloud-based IoMT: see cloud-based
348 IoMT
SVM classifier 334 CNNs 264–6
command line interface (CLI) 362 cognitive rehabilitation, stress
computational intelligence (CI) 34–5, detection for: see cognitive
51 rehabilitation, stress detection for
classification model 37–8 CXRs 263
infections and illnesses, future deadly SARS virus 62
detection of 39 disease, symptoms of 2
machine intelligence, forms of 37 distributed data storage 69
practical case of healthcare provider 69
confusion matrix 49–50 IoT network and data collection 69
framework, workflow of 48–9 IoT technologies: see Internet of
frontal CXR image 48 Things (IoT)
precision-recall curve 50–1 machine learning and deep learning
ROC curves 50 algorithms 263–4, 279–89
Index 447
methodologies in 63–4 selenium 414

RAT 263 vitamin D 413–14
research gaps vitamin E 414
imbalanced datasets 289 zinc 413
large datasets, unavailability of natural antiviral therapies against
289 COVID-19 415
multiple image sources 290 proinflammatory cytokines 408
RT-PCR tests 263 SARS-CoV-2 transmission 406
smart diagnostic/therapeutic COVID-19, cloud-based IoMT for
framework 69–70 architecture of 120
social distancing technologies: see daily-confirmed cases 129
social distancing technologies dataset description 129–30
spread 2 detection techniques
symptoms 63, 262 CT scan 119
touristic destinations, social health imaging modalities 118–19
protection in: see touristic RT-PCR test 118
destinations, social health tests, categories of 117
protection in tests, classes of 117–18
transfer learning thorax CT 118
AlexNet 275 X-ray imaging 119
DenseNet 271–2 digital monitoring tools 135–6
ensemble model 275–6 machine learning and deep learning
generative networks 276 127–9, 137
GoogLeNet 267 methodology
GRNN and PNN 276–7 EfficientNet-B0 architecture 132–3
machine learning 277–9 performance metrics 134–5
MobileNetV2 274 preprocessing 130–2
ResNet 269–71 proposed model 130, 132
SqueezeNet 272–3 ROC curve 134–5
StackNet 266 training model 133–4
VGG networks 267–9 remote monitoring procedures
Xception architecture 273–4 early diagnosis 125–6
YOLU 277 post recovery 127
treatment approaches 64 quarantine time 126–7
COVID-19 antiviral therapies RHM 120–1
CRP 406–7 smartphone applications 125
immune system and inflammatory using, limitations of 136
responds 407–8 wearables 122
immunity, dietary ingredients in 415 drones 124
immunomodulators and adjunctive IoT buttons 124
therapies 409 robots 123
ascorbic acid (vitamin C) 413 smart glasses 122–3
melatonin 412–13 smart helmet 122–3
omega-3 fatty acids 414 smart thermometers 122
phenolic compounds 409–12 WHOOP strap 124–5
COVID-19, risk assessment and spread decision stump 72

of 422, 438–9 decision table 72
components required 436–7 deep learning (DL) 116, 263–4, 433
COVID-19, impact of COVID-19 diagnosis 127–9
agriculture sector 434 state-of-the-art machine learning
economy 434 and, comparison of 279–89
educational sector 435 Dense Convolutional Networ: see
epidemics, prevention and DenseNet
management of 431–2 DenseNet 271–2
sentimental analysis using digital contract tracings (DCT) 16
machine learning 433–4 discrete wavelet transform (DWT)
strategies proposed 432–3 332, 341
general methods 427–8 disruptive digital technologies
fault-tree analysis 429–30 (DDTs) 107–8
what-if analysis 428–9 DL: see deep learning (DL)
WHO 430–1 DWT: see discrete wavelet transform
prediction techniques 426–7 (DWT)
proposed method and working
435–6 ECG: see electrocardiography (ECG)
software required and simulation EDMS: see Electronic Document
437–8 Management System (EDMS)
technology and epidemics E-documents 198–9
dashboard technology 423 EEG: see electroencephalogram (EEG)
Ebola outbreak reaction 422–3 e-Health 214
education 425 EHRs: see electronic health records
hardware 424 (EHRs)
healthcare 425 electrocardiography (ECG) 367
mix use 424–5 electroencephalogram (EEG) 332,
OraQuick 423 340, 367
software 424 electromyography (EMG) 367
work 425 Electronic Document Management
C-reactive protein (CRP) 406, 413 System (EDMS) 198–9
critical literature review (CLR) 87 electronic health records (EHRs) 40
CRP: see C-reactive protein (CRP) electronic medical records (EMRs) 3,
CT scan: see computerized 17
tomography (CT) scan ELM: see extreme learning model
curcumin 411–12 (ELM)
CWT: see continuous wavelet EMG: see electromyography (EMG)
transform (CWT) EMRs: see electronic medical records
CXR: see chest X-ray (CXR) (EMRs)
extreme learning model (ELM) 272
DCT: see digital contract tracings
(DCT) far-UVC lamps 373
DDTs: see disruptive digital fighting COVID-19 pandemic
technologies (DDTs) BDAs
Index 449
big data implementation, process generalized regression neural network

of 40–1 (GRNN) 276–7
computational approaches 40 General Public License (GPL) 186,
COVID-19 outbreak, control and 362
monitoring of 42 Generative Adversarial Networks
drugs manufacturing models, (GAN) 289
production of 44 Geographic Information Systems
environmental regions 42 (GIS) 43
evidence-based actions 40 GIS: see Geographic Information
guiding principle, techniques of 43 Systems (GIS)
integrated big data conceptual global economics and trades,
model 41 COVID-19 on 108–9
outbreak reports 42–3 chances financial support
using, limitations of 44–8 communication sector, CE in
vaccine, development of 43 106–7
vaccine research 43–4 digitalization 107–8
CI 34–5 manage hospital medical and
classification model 37–8 general waste 104–6
infections and illnesses, future circular economy, role of 87–9,
detection of 39 100–1
machine intelligence, forms of 37 categories 103
practical case of 48–51 climate collapse 101–2
supervised learning 38 comprehensive environmental
types of 35 system design 101
unsupervised learning 38–9 social finance system 102–3
using, limitations of 44–8 CLR 87
outbreak 34 framework for
FLW: see food loss and waste (FLW) assessment strategy
food loss and waste (FLW) 390 constructions 89
FTDI: see Future Technology Devices aviation industry 94–6
International Limited (FTDI) CO2 emissions, record decrease
Future Technology Devices in 98–9
International Limited (FTDI) digitalization, rise in 99–100
361 econometric effects 90–1
international commercial trading,
GAN: see Generative Adversarial impact of 94
Networks (GAN) macro-finance impacts 89–90
Gaussian naı̈ve Bayes (GNB) 266, nonpositive impacts of 92–4
278–9 primary energy usage, significant
GDP: see gross domestic product reduction in 98
(GDP) travel sector, significant collision
GDPR: see General Data Protection on 96–8
Regulation (GDPR) social impacts on finance 88–9
General Data Protection Regulation GNB: see Gaussian naı̈ve Bayes
(GDPR) 155, 172, 222 (GNB)
Google Meet 201–2 network 5

GoogLeNet 267 security 6
GPL: see General Public License things or devices 4
(GPL) user interface 5
GRNN: see generalized regression confront COVID-19
neural network (GRNN) aspects of 9–10
gross domestic product (GDP) 92 disinfecting area 13
health monitoring systems 10–12
HADDOCK: see high ambiguity– logistics delivery 14
driven protein–protein telemedicine 14
DOCKing (HADDOCK) tracking and detecting possible
HealthyBlock 162 patients 12–13
heart rate variability (HRV) 125 contributions of 3
high ambiguity–driven protein–protein data collection 4
DOCKing (HADDOCK) 313 data management 4
histogram of oriented gradient (HOG) data transmission 4
48 low-cost technologies 4
HLA: see human leukocyte antiques possible solutions and deliverables
(HLA) 19–20
HOG: see histogram of oriented IoMT: see Internet of Medical Things
gradient (HOG) (IoMT)
HRV: see heart rate variability (HRV) iterative neighbourhood component
human leukocyte antiques (HLA) 303 analysis (INCA) 277
ICTs: see information and K-nearest neighbours (K-NN) 72, 266

communication technologies K-NN: see K-nearest neighbours
(ICTs) (K-NN)
image augmentation operation 134
inactivated vaccine 65 learning management systems
INCA: see iterative neighbourhood (LMSs) 183–6, 204
component analysis (INCA) LED: see light emitting diode (LED)
information and communication Lesser General Public License
technologies (ICTs) 34 (LGPL) 362
Internet of Medical Things (IoMT) 3, LGPL: see Lesser General Public
116 License (LGPL)
blockchain technologies 18 LIDARs: see light detection and
COVID-19, cloud-based IoMT for: ranging (LIDARs)
see COVID-19, cloud-based light detection and ranging (LIDARs)
IoMT for 379
Internet of Things (IoT) 20–2 light emitting diode (LED) 376
challenges of 18–19 LINE 191, 200
components of 4–5 live-attenuated vaccine 65
analytics 5 LMSs: see learning management
central control 6 systems (LMSs)
cloud 5 long short-term memory (LSTM) 335
Index 451
LSTM: see long short-term memory MLP: see multi-layer perceptron (MLP)
(LSTM) MobileNetV2 274
lycorine 317 motor driver 374
MQTT: see message queuing telemetry
machine-assisted designing and transport (MQTT)
evaluation of COVID-19 drug MRF model: see Markov random field
high ambiguity–driven protein– (MRF) model
protein DOCKing 313–16 mRNA: see messenger RNA (mRNA)
molecular docking program MSW: see municipal solid waste
curcumin 313 (MSW)
favipiravir 312 multi-layer perceptron (MLP) 268
montelukast 313 municipal solid waste (MSW) 386
M-pro enzyme 312
N4-QO molecule 310 Naı̈ve Bayes 71–2
natural and synthetic drugs 309 nidovirus RdRc-associated
nelfinavir 313 nucleotidyltransferase
remdesivir 312 (NiRAN) 299
molecular dynamics 316–18 NiRAN: see nidovirus RdRc-
machine learning (ML) 37, 116, 182, associated
263–4, 277–9, 331 nucleotidyltransferase
COVID-19 diagnosis 127–9 (NiRAN)
IoT-based framework 70–2 NodeMCU: see node micro controller
sentimental analysis using 433–4 unit (NodeMCU)
state-of-the-art deep learning and, node micro controller unit
comparison of 279–89 (NodeMCU) 369
management information system
(MIS) 199–200 omega-3 fatty acids 414
Markov random field (MRF) model 229 OneR: see One rule (OneR)
mathematical model 427 One rule (OneR) 72
MDH: see medical diagnosis humanoid
(MDH) PCA: see principal component analysis
medical diagnosis humanoid (MDH) 365 (PCA)
melatonin 412–13 PCR: see polymerase chain reaction
mental stress 332 (PCR)
MERS: see Middle East respiratory personal protective equipment (PPE) 386
syndrome (MERS) PNN: see probabilistic neural network
message queuing telemetry transport (PNN)
(MQTT) 365 polymerase chain reaction (PCR) 63
messenger RNA (mRNA) 297 PPE: see personal protective
microarray data analysis 233, 244–6 equipment (PPE)
Middle East respiratory syndrome principal component analysis (PCA)
(MERS) 387 39, 334, 342, 346–7
MIS: see management information probabilistic neural network (PNN)
system (MIS) 276–7
ML: see machine learning (ML) propolis 412
QoS: see quality of service (QoS) SARS: see severe acute respiratory
QTs: see quantitative traits (QTs) syndrome (SARS)
quality of service (QoS) 367 SARS-CoV-2: see severe acute
quantitative traits (QTs) 235 respiratory syndrome
quercetin 412 coronavirus-2 (SARS-CoV-2)
scRNA-seq data analysis 234–5,
radial basis function (RBF) 335 239–40
radio frequency identification (RFID) SDGs: see Sustainable Development
366 Goals (SDGs)
rapid antibody testing (RAT) 262 SDN: see software development
RAT: see rapid antibody testing (RAT) network (SDN)
RBD: see receptor-binding domain selenium 414
(RBD) severe acute respiratory syndrome
RBF: see radial basis function (RBF) coronavirus-2 (SARS-CoV-2)
real-time quantitative PCR 116, 324–5, 406
(RT-qPCR) 305 ACE2 enzyme 301
receptor-binding domain (RBD) 299 amino acids 323
recursion-based prediction COVID-19 chromosomes 298
model 427 genetic laboratory analysis reports
rehabilitation 333 297
remote-healthcare-monitoring (RHM) in human 300
120 machine-integrated advanced
ResNet 269–71 techniques
respiratory rate 125 computerized tomography 318–19
resting heart rate (RHR) 125 COVID-19 treatment, MRI for
RESV: see resveratrol (RESV) 320–2
resveratrol (RESV) 411 microscopic and 3D molecular
reverse transcription polymerase chain image of 298
reaction (RT-PCR) 262 NiRAN 299–300
RFID: see radio frequency physiopathogenesis of 407
identification (RFID) pp1ab and pp1a 299
RHM: see remote-healthcare- real-time COVID-19 identification
monitoring (RHM) test (RT-PCR)
RHR: see resting heart rate (RHR) isothermal enhancement 304
ribonucleic acid (RNA) 297 limitations of 304–5
RNA: see ribonucleic acid (RNA) nuclear chain antiques 304
RNA-seq: see RNA sequencing protein-based findings 304
(RNA-seq) ribosome 300
RNA-seq data analysis 233–4, 241–3 silico crystallographic results 301
RNA sequencing (RNA-seq) 231 silico methods
RT-PCR: see reverse transcription assisted anchoring site analysis
polymerase chain reaction 305–7
(RT-PCR) machine-assisted designing and
RT-qPCR: see real-time quantitative evaluation of COVID-19 drug
PCR (RT-qPCR) 308–18
Index 453
TMPRSS2 enzyme 301 manufacturing

vaccine, human antibody generation distance monitoring using video
and role of analytics 194
COVID-19, synthetic vaccine on distance monitoring using Wi-Fi
302–3 194
immediate action of 302 distance using wearable device
vaccines 323 193–4
severe acute respiratory syndrome robots 195
(SARS) 387 methodology 182
silico methods in COVID-19 drug and smart city
research AI and big data 202
assisted anchoring site analysis implementation and usability
AutoDock Vina 305–6 202–3
DockThor-VS platform 307 policy and legislation 203
molecular dynamics 306–7 privacy and security 203
machine-assisted designing and social networking and conference
evaluation of COVID-19 drug software for, education 186–7
high ambiguity–driven protein– supporting everyday life
protein DOCKing 313–16 applications support WFH
molecular docking program service 196–200
308–13, 323 conferencing application 200–2
molecular dynamics 316–18, 323 remote working 196–8
SIR model: see susceptible recovered software development network (SDN)
infectious (SIR) model 170
SLR: see systematic literature review SqueezeNet 272–3
(SLR) StackNet 266
smartphones, COVID-19 monitoring standard logistic regression model 164
using 136 subunit approach, COVID-19
social contact tracing 135–6 pandemic 66
social distancing technologies 181–2, genetic approach (nucleic acid
203–5 vaccine) 66
healthcare timeline 66–8
MorChana application 191–2 support vector machine (SVM) 71,
payment system 189–91 266, 332
queue systems 188–9 susceptible recovered infectious (SIR)
screening system 188 model 163–4
ThaiChana application 192 Sustainable Development Goals
tracking infection information (SDGs) 96
191–2 SVM: see support vector machine
wearable technology 187–8 (SVM)
learning management systems systematic literature review (SLR) 87
(LMSs) 204
Google Classroom 184–5 ThaiChana application 192
MOOCs 183–4 time-to-event analytics model 164–5
Moodle 185–6 TLRs: see toll-like receptors (TLRs)
toll-like receptors (TLRs) 302 data management and learning

touristic destinations, social health 396
protection in 222–3 fund raising and national and
audiovisual communication 213 international collaboration 397
cloud platform 214 infectious waste, handling of 395
data protection 220–2 information, education, and
health protection, proposed software communication 396
solution for managing waste, basic principles
architecture of 215–16 for 396–7
border control 220 policy, regulatory, and
healthcare service 217–18 guidelines 395
local government service 219–20 segregation, monitoring of 396
system architecture 215–17 suitable disposal methods 396
tourist service 218–19 socioeconomic and environmental
implemented information and impacts 387–8
communication technology waste generation, impact of
system 213 dead bodies 393
IoT technology, task of 213 food waste 390
mobile application 212 good waste management 391
people’s daily activities 210, 222 healthcare waste/BMW 392
research 214 household waste/MSW 392–3
restrictions 210 mask used data 390
telemedicine, use of 212 plastic waste 390
PPE waste 389–90, 393
UAV: see uncrewed aerial vehicle (UAV) wastewater control approach
ultrasonic sensors 372–3 388–9
ultrasound 262 world, regions of 389
uncrewed aerial vehicle (UAV) 124 WBAN: see wireless body area
network (WBAN)
12-V battery 374–5 Webex 202
VGG networks 267–9 WHO: see World Health Organization
viral vector vaccine 65 (WHO)
virtual private network (VPN) 196 whole-microbe approach 65
vitamin D 413–14 wireless body area network (WBAN)
vitamin E 414 367
VPN: see virtual private network (VPN) World Health Organization (WHO) 86
Wanda QuickTouch 124 Xception architecture 273–4

waste management system, COVID-19
pandemic on 397–8 YOLU 277
BMW management 386, 392
categories 386 ZeroR: see Zero rule (ZeroR)
handling waste, challenges in 393–4 Zero rule (ZeroR) 72
rethinking effective waste zinc 413
management 394–5 Zoom applications 201

Untitled

Uploaded by

Copyright:

Available Formats

Untitled

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Untitled

Uploaded by

Copyright:

Available Formats

HEALTHCARE TECHNOLOGIES SERIES 42

Could you be our next author?

Proposals for coherently integrated international multi-authored edited or co-authored

The Institution of Engineering and Technology

The Institution of Engineering and Technology

British Library Cataloguing in Publication Data

ISBN 978-1-83953-518-5 (hardback)

Typeset in India by MPS Limited

About the editors xv

1 Internet of Things (IoT) and blockchain-based solutions to confront

2 Application of big data and computational intelligence in fighting

2.2 Applicability of computational intelligence in combating

3 Cloud-based IoMT for early COVID-19 diagnosis and monitoring 61

4 Assessment analysis of COVID-19 on the global economics

4.4 Framework for the international financial system,

5 Early diagnosis and remote monitoring using cloud-based

5.10 Limitations of using IoMT devices 136

6 Blockchain technology for secure COVID-19 pandemic

7 Social distancing technologies for COVID-19 181

7.3.1 Learning management system 183

8 Social health protection in touristic destinations during COVID-19 209

9 Analysis of Artificial Intelligence and Internet of Things in

9.2 Definition of biomedical keywords 230

10 Review of medical imaging with machine learning and deep

11 Machine-based drug design to inhibit SARS-CoV-2 virus 295

11.3 Mechanism of SARS-coronavirus-2 infection in human 299

12 Stress detection for cognitive rehabilitation in

13 Arduino-based robot for purification of COVID-19 using far

14 Effect of COVID-19 pandemic on waste management system

15 Natural adjunctive therapies options other than COVID-19

15.4.2 Melatonin 412

16 Risk assessment and spread of COVID-19 421

Chinmay Chakraborty, Birla Institute of Technology, India

Chinmay Chakraborty has published more than 70 conference presentations, jour-

Joel J.P.C. Rodrigues, College of Computer Science and Technology, China

Chapter 1: Internet of Things (IoT) and blockchain-based

Chapter 2: Application of big data and computational

Chapter 3: Cloud-based IoMT for early COVID-19

Chapter 4: Assessment analysis of COVID-19 on the global

advocates for a complete overhaul of the global economic development paradigm,

Chapter 5: Early diagnosis and remote monitoring using

Chapter 6: Blockchain technology for secure COVID-19

Chapter 7: Social-distancing technologies for COVID-19

Chapter 8: Social health protection in touristic destinations

Chapter 9: Analysis of Artificial Intelligence and Internet of

Chapter 10: Review of medical imaging with machine

Chapter 11: Machine-based drug design to inhibit SARS-

Chapter 12: Stress detection for cognitive rehabilitation in

Chapter 13: Arduino-based robot for purification of

Chapter 14: Effect of COVID-19 pandemic on waste

Chapter 15: Natural adjunctive therapies options other than

Chapter 16: Risk assessment and spread of COVID-19

COVID-19 pandemic, an unprecedented event that has severely affected every

monitoring, secure data handling, and telemedicine. Finally, challenges of using

Keywords: COVID-19; IoT; Blockchain; Contact tracing; Telemedicine;

“Coronavirus Disease 2019” that is mostly known as “COVID-19” is mainly a

● We provided an overview of the main components of IoT and blockchain

1.2 Internet of Things (IoT) and blockchain overview

This section contains an overview of IoT and blockchain technology.

1.2.1 Internet of Things