Conference Presentations by Nandana Pathirage
10th International Research Conference - General Sir John Kotelawala Defence University, Sri Lanka, 2017
Leukemia, simply called "Blood cancer" is a fatal disease where the white blood cells (WBC) incre... more Leukemia, simply called "Blood cancer" is a fatal disease where the white blood cells (WBC) increases in bone marrow and peripheral blood. Acute lymphoblastic leukemia (ALL) is one of the most common types of leukemia aroused by accumulation and overproduction of immature and cancerous cells known as lymphoblasts. Presently, the diagnosis of ALL includes performing a full blood count, blood picture, bone marrow biopsy, cytochemical stain, immunophenotyping and cytogenetics. These techniques are highly tedious, costly, requires expertise of hematologists and available only in few hospitals. Therefore, as an alternative, use of image processing to diagnose ALL would become an effective solution. Although, several research groups have employed image processing to identify ALL, recognition and splitting of overlapping red blood cells (RBC) with WBC has yet been a challenging issue. This paper is about an application which includes an image processing algorithm to diagnose ALL while attempting to solve the above mentioned issue of overlapping cells. The algorithm is also extended to detect the quality devastation in blood films in terms of storing them for prolonged period. The inputs for this application are microscopic peripheral blood images of ALL patients obtained from Department of Pathology Clinic at Faculty of Medicine, University of Colombo. Then, image processing techniques; image enhancement, segmentation, feature extraction and classification are performed. For the detection and diagnosis of leukemia, segmentation using morphological operations in OpenCV Python and classification using K-Nearest Neighbour and Support vector machine implementations has been proposed in this research. It is observed that the proposed algorithm has led to a high accuracy in diagnosing ALL. The system also includes a PHP based web application that serves hematologists, doctors and patients to log in to their specific user accounts and make records, insert details and view diagnosing reports.
Papers by Nandana Pathirage
International journal on advances in ICT for emerging regions, Dec 17, 2023
International Journal on Digital Libraries, Jan 11, 2024
International Journal on Digital Libraries
Reading list systems are widely used in tertiary education as a pedagogical tool and for tracking... more Reading list systems are widely used in tertiary education as a pedagogical tool and for tracking copyrighted material. This article explores the make-up of reading lists across a whole university. We investigated the experience of academics and librarians when creating reading lists. A mixed-method approach was employed in which we performed a transaction log analysis on reading lists at a single university, from 2016 to 2020. A questionnaire was then answered by both academics and academic liaison librarians about their experience with reading lists. The results of our analysis found that uptake of reading lists varies widely between different academic disciplines. Academic engagement with reading lists was found to show only incremental growth over time, and overall satisfaction by academics with the reading list system was low. We explore implications for reading lists implemented through digital Libraries and recommend developing discipline-specific support to increase reading ...
2022 International Conference on Emerging Techniques in Computational Intelligence (ICETCI)
Leukemia, simply called “Blood cancer” is a fatal disease where the white blood cells (WBC) incre... more Leukemia, simply called “Blood cancer” is a fatal disease where the white blood cells (WBC) increases in bone marrow and peripheral blood. Acute lymphoblastic leukemia (ALL) is one of the most common types of leukemia aroused by accumulation and overproduction of immature and cancerous cells known as lymphoblasts. Presently, the diagnosis of ALL includes performing a full blood count, blood picture, bone marrow biopsy, cytochemical stain, immunophenotyping and cytogenetics. These techniques are highly tedious, costly, requires expertise of hematologists and available only in few hospitals. Therefore, as an alternative, use of image processing to diagnose ALL would become an effective solution. Although, several research groups have employed image processing to identify ALL, recognition and splitting of overlapping red blood cells (RBC) with WBC has yet been a challenging issue. This paper is about an application which includes an image processing algorithm to diagnose ALL while atte...
2020 20th International Conference on Advances in ICT for Emerging Regions (ICTer), 2020
Fake news is a new phenomenon related to false information and fraud that spreads through online ... more Fake news is a new phenomenon related to false information and fraud that spreads through online social media or traditional news media. Today, fake news can be easily created and distributed across many social media platforms and has a widespread impact on the real world. It is critical to develop efficient algorithms and tools for early detection of how false information is disseminated on social media platforms and why it is successful in deceiving users. Most research methods today are based on machine learning, deep learning, feature engineering, graph mining, image and video analysis and newly developed datasets and web services for detecting deceptive content. Therefore, a strong need emerges to find a suitable method that can easily detect false information. A hybrid approach has suggested using the CNN model and RNN-LSTM model to detect false information from this study. First, NLTK toolkit has used to remove stop words, punctuations and special characters from the text. Then the same toolkit applies to tokenize the text and preprocesses the text. From there on, GloVe word embeddings have added to the preprocessed text. Higher-level features of the input text extract from the CNN model using convolutional layers and max-pooling layers. Long-term dependencies between word sequences capture from RNN-LSTM model. The suggested model also applies dropout technology with Dense layers to enhance the efficiency of the hybrid model. Results of the suggested hybrid model have shown that the suggested CNN, RNN-LSTM based Hybrid approach achieves the highest accuracy of 92% by surpassing most of the classical models today with Adam optimizer and Binary Cross-Entropy loss function.
2021 21st International Conference on Advances in ICT for Emerging Regions (ICter)
Leukemia, simply called “Blood cancer” is a fatal disease where the white blood cells (WBC) incre... more Leukemia, simply called “Blood cancer” is a fatal disease where the white blood cells (WBC) increases in bone marrow and peripheral blood. Acute lymphoblastic leukemia (ALL) is one of the most common types of leukemia aroused by accumulation and overproduction of immature and cancerous cells known as lymphoblasts. Presently, the diagnosis of ALL includes performing a full blood count, blood picture, bone marrow biopsy, cytochemical stain, immunophenotyping and cytogenetics. These techniques are highly tedious, costly, requires expertise of hematologists and available only in few hospitals. Therefore, as an alternative, use of image processing to diagnose ALL would become an effective solution. Although, several research groups have employed image processing to identify ALL, recognition and splitting of overlapping red blood cells (RBC) with WBC has yet been a challenging issue. This paper is about an application which includes an image processing algorithm to diagnose ALL while atte...
Intensive care units(ICU) in Sri Lankan hospital systems are generally managed manually. Using a ... more Intensive care units(ICU) in Sri Lankan hospital systems are generally managed manually. Using a manual system for the ICU can cause many difficulties as the ICU is the main department in a hospital. After analysing the current procedure at Kalubowila Teaching Hospital we observed that it normally takes 24 hours to identify and direct a seriously ill dengue patient to the ICU from the High Dependency Unit(HDU). Patients with the Dengue virus will have similar symptoms. Thereby the next patient who needs to be admitted into the ICU has to be more ailing than the rest of the patients in the HDU. If the correct patient to be admitted is not identified, another patient in the unit could fall into a more severe case. ICU Management System is a web based system that has the ability to identify the next most suitable Dengue patient that should be treated in the ICU using a specific score. The score is calculated by monitoring the symptoms of the patient and giving a separate value to each ...
Leukemia is a fatal disease of the type “Blood Cancer”, where the White Blood Cells (WBC) increas... more Leukemia is a fatal disease of the type “Blood Cancer”, where the White Blood Cells (WBC) increases in human bone marrow and peripheral blood. Acute Lymphoblastic Leukemia (ALL) is a common types of leukemia that affects young children of below 10 years and adults over 60 years, aroused by accumulation and overproduction of immature and cancerous cells identified as lymphoblasts. At present, the diagnosis of ALL includes measures alike performing a full blood count, bone marrow biopsy, blood picture, immunophenotyping, cytochemical stain and cytogenetics. These medicinal techniques are highly tedious, costly, requires expertise of hematologists and available only in few hospitals especially in developing countries. Hence, as an alternative, use of image processing and machine learning to diagnose ALL would become an effective solution. Even though, several research groups have used image processing to detect and diagnose ALL, recognition and splitting of overlapping Red Blood Cells ...
Intensive care units(ICU) in Sri Lankan hospital systems are generally managed manually. Using a ... more Intensive care units(ICU) in Sri Lankan hospital systems are generally managed manually. Using a manual system for the ICU can cause many difficulties as the ICU is the main department in a hospital. After analysing the current procedure at Kalubowila Teaching Hospital we observed that it normally takes 24 hours to identify and direct a seriously ill dengue patient to the ICU from the High Dependency Unit(HDU). Patients with the Dengue virus will have similar symptoms. Thereby the next patient who needs to be admitted into the ICU has to be more ailing than the rest of the patients in the HDU. If the correct patient to be admitted is not identified, another patient in the unit could fall into a more severe case. ICU Management System is a web based system that has the ability to identify the next most suitable Dengue patient that should be treated in the ICU using a specific score. The score is calculated by monitoring the symptoms of the patient and giving a separate value to each ...
Inaccurate range estimation is a major problem which comes with Electric Vehicles. Because of thi... more Inaccurate range estimation is a major problem which comes with Electric Vehicles. Because of this many people face issues when planning long trips and short trips with limited battery capacity. To overcome this issue, it is necessary to have a better power consumption prediction algorithm which uses vehicle data and other dynamic environmental conditions. This paper is based on cloud based power consumption estimation system which uses linear regression in machine learning to obtain a better estimation based on above mentioned areas. KeywordsElectric Vehicle, Range, Power Consumption, Estimation
Diabetes is one of deadliest diseases in the world. As per the existing system in Sri Lanka, pati... more Diabetes is one of deadliest diseases in the world. As per the existing system in Sri Lanka, patients have to visit a diagnostic center, consult their doctor and wait for a day or more to get their result. Moreover, every time they want to get their diagnosis report, they have to waste their money in vain. But with the rise of Machine Learning approaches, we have been able to find a solution to this problem using data mining. Data mining is one of the key areas of Machine learning. It plays a significant role in diabetes research because It has the ability to extract hidden knowledge from a huge amount of diabetes related data. The aim of this research is to develop a system which can predict whether the patient has diabetes or not. Furthermore, predicting the disease early leads to treatment of the patients before it becomes critical. This research has focused on developing a system based on three classification methods namely, Decision Tree, Naïve Bayes and Support Vector Machine ...
2021 International Research Conference on Smart Computing and Systems Engineering (SCSE)
Journal of Theoretical and Applied Information Technology (JATIT), Sep 30, 2020
Leukemia is a fatal disease of the type "Blood Cancer", where the White Blood Cells (WBC) increas... more Leukemia is a fatal disease of the type "Blood Cancer", where the White Blood Cells (WBC) increases in human bone marrow and peripheral blood. Acute Lymphoblastic Leukemia (ALL) is a common types of leukemia that affects young children of below 10 years and adults over 60 years, aroused by accumulation and overproduction of immature and cancerous cells identified as lymphoblasts. At present, the diagnosis of ALL includes measures alike performing a full blood count, bone marrow biopsy, blood picture, immunophenotyping, cytochemical stain and cytogenetics. These medicinal techniques are highly tedious, costly, requires expertise of hematologists and available only in few hospitals especially in developing countries. Hence, as an alternative, use of image processing and machine learning to diagnose ALL would become an effective solution. Even though, several research groups have used image processing to detect and diagnose ALL, recognition and splitting of overlapping Red Blood Cells (RBC) with WBC has however been a challenging issue. This paper is about a research study and an application that includes an image processing and machine learning algorithm to diagnose ALL while attempting to solve the issue of overlapping cells. The research is also extended to detect the quality devastation in blood films in terms of storing them for prolonged period. The inputs for this application include microscopic peripheral blood films of ALL patients and healthy individuals obtained Lanka. This research project has received verification of ethical approval from Faculty of Medicine, General Sir John Kotelawala Defence University, Sri Lanka. In the developed application, segmentation using morphological operations in OpenCV Python and supervised learning based classification using K-Nearest Neighbour implementation has been proposed in detection and diagnosing of ALL. As per the results, the proposed algorithm has led to a high accuracy of 88.8% in diagnosing ALL. The end product includes a Python based QT GUI based development suite that performs main targeted backend functionalities and a PHP based web application that serves hematologists, doctors and patients to perform utility functions.
Uploads
Conference Presentations by Nandana Pathirage
Papers by Nandana Pathirage