Uvais Qidwai
Qatar University, Computer Science and Engineering, Faculty Member
Children with autism face challenges in various skills (e.g., communication and social) and they exhibit challenging behaviours. These challenging behaviours represent a challenge to their families, therapists, and caregivers, especially... more
Children with autism face challenges in various skills (e.g., communication and social) and they exhibit challenging behaviours. These challenging behaviours represent a challenge to their families, therapists, and caregivers, especially during therapy sessions. In this study, we have investigated several machine learning techniques and data modalities acquired using wearable sensors from children with autism during their interactions with social robots and toys in their potential to detect challenging behaviours. Each child wore a wearable device that collected data. Video annotations of the sessions were used to identify the occurrence of challenging behaviours. Extracted time features (i.e., mean, standard deviation, min, and max) in conjunction with four machine learning techniques were considered to detect challenging behaviors. The heart rate variability (HRV) changes have also been investigated in this study. The XGBoost algorithm has achieved the best performance (i.e., an a...
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Stroke patients, as well as all those patients who are bed-bound for a long period of time, are highly susceptible to preventable secondary complications such as pressure ulcers or commonly known as bedsores. Such secondary complications... more
Stroke patients, as well as all those patients who are bed-bound for a long period of time, are highly susceptible to preventable secondary complications such as pressure ulcers or commonly known as bedsores. Such secondary complications may lead to progression of symptoms and are an important cause of delayed hospital discharges or even fatalities. The unnecessary long occupancy of the bed causes stress on the hospital's operations as well and adds to the operational costs. Newer strategies are, therefore, urgently needed to improve detection of patients at risk and to prevent such complications. In this paper, a system has been presented for detection and quantification of patient's position as a measure of possible critical triggering scenario that can lead to the development of the pressure ulcers. This multidisciplinary work combines the clinical knowledge of physicians and scientists at the Stroke ward with the instrumentation and intelligent algorithm into an add-on system that can be easily attached to the existing beds in any typical neurological clinics as well as intensive care units (ICU). Data fusion is assessed continuously with intelligent algorithms to alert the medical staff about patient's physical condition, e.g. turning to reduce the risk of post-stroke complications. The built device was tested with healthy volunteers and a detailed study is being conducted at the hospital to evaluate the clinical feasibility of the design.
Research Interests:
Detection of mitotic cells in histology images is an important but challenging process due to the resemblance of mitotic cells with other non-mitotic cells and also due to the different appearance of mitotic cells undergoing different... more
Detection of mitotic cells in histology images is an important but challenging process due to the resemblance of mitotic cells with other non-mitotic cells and also due to the different appearance of mitotic cells undergoing different phases of the division process. In this paper, we present an algorithm for classification of mitotic cells into its four different phases using eigenphase nuclei images - nuclear exemplars obtained separately from the eigen-decomposition of training nuclei images belonging to each of the four mitotic phases. To the best of our knowledge, ours is the first method to identify mitotic phases in cancer histology images. It is quite likely that the classification results may be negatively affected if the dataset used for training purposes does not contain sufficient number of samples for a positive class. To overcome this class imbalance problem, we present a novel method for oversampling the minority class. The proposed method generates synthetic images for training purposes by perturbing the representation of training samples belonging to the minority class in the eigenphase domain. We show that this strategy works effectively for pairwise classification of the mitotic cells - increasing the classification performance by as much as 24%.
Research Interests:
Subspace identification is considered to be one of the most powerful techniques in the system identification domain. In this paper, an application of a powerful version of subspace identification, N4SID, is presented. The target area is... more
Subspace identification is considered to be one of the most powerful techniques in the system identification domain. In this paper, an application of a powerful version of subspace identification, N4SID, is presented. The target area is ultrasonic A-scans for defect detection in metals. Simulated as well as real data sets have been used for testing. Since the A-scans are a very commonly used form of analysis signals for nondestructive testing (NDT), their analysis is very useful for determining the type and severity of defect.
Research Interests:
Electroencephalography (EEG) plays an intelligent role, especially EEG based health diagnosis of brain disorder, as well as brain-computer interface (BCI) applications. One such research field is related to epilepsy. The EEG based methods... more
Electroencephalography (EEG) plays an intelligent role, especially EEG based health diagnosis of brain disorder, as well as brain-computer interface (BCI) applications. One such research field is related to epilepsy. The EEG based methods are not will designed for pre-occurrence recognition scheme to detect and predict partial seizure for epileptic patients. The system even becomes more complicated if the detection system is to be designed for ubiquitous operations, for the identification of people with seizure disabilities. In this case, the patients are not restricted to the clinical environment in which many devices are involved to the patient externally while he/she can continue daily activities. This paper demonstrates a classification method by using Fuzzy Logic System to identify, predict the Partial Seizure from Epileptic data. Here the paper shows preliminary results of the normal state, pre-seizure state and seizure state of the subject's brain signal data. This can be observed and the algorithm with the detection structure can produce cautioning signals for epileptic seizure.
Research Interests: Computer Science, Fuzzy Logic, Embedded Systems, Epilepsy, Electroencephalography, and 10 moreBrain Computer Interface, Fuzzy Systems, Databases, Brain Computer Interfaces, Real Time Systems, Wearable computers, Knowledge Based Systems, Wearable Computer, Classification Algorithms, and Rule Based System
In this paper, infrared (IR) imaging technique is used in conjunction with image deconvolution algorithms in order to enhance detection capabilities of the NDT personnel in detecting defects such as cracks, voids, and pitting. The main... more
In this paper, infrared (IR) imaging technique is used in conjunction with image deconvolution algorithms in order to enhance detection capabilities of the NDT personnel in detecting defects such as cracks, voids, and pitting. The main idea utilized here is the fact that the heat distribution remains fairly homogenous in a healthy metal and can be seen as a flat surface under IR imaging camera. However, things change drastically if there is a discontinuity in the homogenous metal, such as defects. This change is exploited in this work as an output of a defect function which produces this output when healthy metal image is given as an input to it. H¿ deconvolution methodology has been utilized here to isolate (deconvolve) the defect function and be able to regenerate the output image without any other obscuring elements. This can enhance the detection capability of the IR camera as well as the NDT personnel.
Research Interests:
In this paper, an inspection system has been presented that uses Laser-Ultrasound (LU) technique for Nondestructive testing (NDT) of metallic structures with specific interest in Oil & Gas sector. While the developed system is... more
In this paper, an inspection system has been presented that uses Laser-Ultrasound (LU) technique for Nondestructive testing (NDT) of metallic structures with specific interest in Oil & Gas sector. While the developed system is the first one of its kind in the Middle-Eastern region, the nature of signals are quite unique as well and traditional signal processing runs into a
Research Interests:
This article shares my experiences with the project-based methodology of teaching the same material that a standard text in robotics would teach and will comparatively elaborate the improvement observed in the group of students involved.... more
This article shares my experiences with the project-based methodology of teaching the same material that a standard text in robotics would teach and will comparatively elaborate the improvement observed in the group of students involved. Other than the understanding of the course material, the use of project-based learning methodology has shown a great increase in motivation and focus on the subject in the students towards the engineering education. The performance improvement is presented in this work as a comparison with respect to underlying program educational outcomes between the previous years of teaching in conventional way, with past two years of experience using the projectbased approach. The work includes appropriate rubrics involved in evaluating the subject. While the work is seemingly undergraduate teaching exercise, it has proven so far to have very positive impact on the industrial research project performed by the author.
Research Interests:
We present a human-centric framework for pattern classification. We call the framework human-centric because the classifier depends on the judgment and prior experience of a human expert for the interpretation of weak learner scores. In... more
We present a human-centric framework for pattern classification. We call the framework human-centric because the classifier depends on the judgment and prior experience of a human expert for the interpretation of weak learner scores. In the first step a large number of simple Fuzzy Inference Engines (FIEs) are constructed to perform classification based on linguistic rules for weak learner score
Research Interests:
... REFERENCES [1] M. Oren, C. Papageorgiou, P. Sinha, E. Osuna, and T. Poggio, Pedestrian detection using wavelet templates, IEEE Computer ... based face detection, IEEE International Conference on Analysis and Modeling of Faces and... more
... REFERENCES [1] M. Oren, C. Papageorgiou, P. Sinha, E. Osuna, and T. Poggio, Pedestrian detection using wavelet templates, IEEE Computer ... based face detection, IEEE International Conference on Analysis and Modeling of Faces and Gestures, 2003 [5] R. Iqbal, and U ...
Research Interests:
ABSTRACT In this paper, an Intelligent Blind-Spot Sensing system is presented. This system exploits the power of sensor technology using multiple low-cost ultrasonic sensors to monitor the designated blind spots in a vehicle. In addition,... more
ABSTRACT In this paper, an Intelligent Blind-Spot Sensing system is presented. This system exploits the power of sensor technology using multiple low-cost ultrasonic sensors to monitor the designated blind spots in a vehicle. In addition, an embedded Basic Stamp controller is used as the brain of the system that controls the actions of ultrasonic sensors as well as produces various warning alarms for the driver. The system uses a Fuzzy Inference System (FIS) as the main intelligence and the main decision logic of this system. Because of low-cost sensors, the readings are not very stable or precise all the time. It's the interpretation of the sensor data through the FIS which is responsible for the correctness of the decision. This system will assist the drivers in safe- maneuvering to prevent accidents. The FIS is structured to incorporate human perceptive power in making a similar decision as to whether there is danger or not..
Research Interests:
ABSTRACT An obstacle avoidance system was developed for a custom-made autonomous navigational robotic vehicle (ANROV), based on an intelligent sensor network and fuzzy logic control. Unlike conventional crisp control systems, this system... more
ABSTRACT An obstacle avoidance system was developed for a custom-made autonomous navigational robotic vehicle (ANROV), based on an intelligent sensor network and fuzzy logic control. Unlike conventional crisp control systems, this system is based on soft computing, such that the system mimics human decision making processes and can better recognize obstacles and decide the best course of action. The fuzzy logic control system was developed in MATLAB, the GUI was designed in LabView, and the hardware realization was installed in ANROV for evaluation and development
Research Interests:
It is desirable to have a monitoring system that can keep a constant surveillance on the conditions of the heart and its related patterns. This is particularly important in many patients with critical cardiac abnormalities. This can be... more
It is desirable to have a monitoring system that can keep a constant surveillance on the conditions of the heart and its related patterns. This is particularly important in many patients with critical cardiac abnormalities. This can be very convenient in clinical settings but may not be possible for individuals who are not in hospital and are in their day-to-day activities. Wearable ECG-based systems have been proposed for such situations and can perform such monitoring in real life. However, detecting the abnormality in near real-time is still a ...
Research Interests: Electrical Engineering, Computer Science, Biomedical Engineering, Ubiquitous Computing, Health Care, and 15 moreEmbedded Systems, Medicine, Healthcare, Identification, Data acquisition, Patient Monitoring, Wearable Computer, WBAN, Fibrillation, Information Security in Healthcare, Android Operating System, Component, Energy Levels, Electrocardiogram (ECG) Signals, and Springer Ebooks
ABSTRACT Knowing the frequency contents present in a network enables the operation and design engineers to predict the functioning of the system with respect to certain equipment and can also be used to identify the location and extent of... more
ABSTRACT Knowing the frequency contents present in a network enables the operation and design engineers to predict the functioning of the system with respect to certain equipment and can also be used to identify the location and extent of use of filters and other protective devices. The usual frequency estimation techniques are dependent on some linear approximation for the nonlinear distortion present in the system. This paper presents the use of Genetic Algorithms (GAs) with nonlinear system structure to estimate the amplitudes and phases of various frequency components present in the system. In this paper, the general nonlinear formulation problem of estimation of harmonics is solved using GAs.
Research Interests:
... M Shahzad Shamim Consultant Neurosurgeon Aga Khan University Hospital Karachi, Pakistan Saniya Siraj Godil Research Officer Aga Khan University Hospital Karachi, Pakistan Mohsin Qadeer Resident Surgeon Aga Khan University Hospital... more
... M Shahzad Shamim Consultant Neurosurgeon Aga Khan University Hospital Karachi, Pakistan Saniya Siraj Godil Research Officer Aga Khan University Hospital Karachi, Pakistan Mohsin Qadeer Resident Surgeon Aga Khan University Hospital Karachi, Pakistan ...
Research Interests:
Research Interests:
... Fig. 6. Fundamental harmonic with multiple frequency estimation technique. View Within Article. Fig. 7 shows the estimation for the amplitude of fifth harmonic. The actual value is 0.09 pu and the phase is 82.1°. In the case of ...
Research Interests:
In this article a new strategy is presented that can be used by neurophysicians, neurosurgeons, and orthopedic surgeons to predict patients' health after an operative procedure on the vertebral column just by analyzing the... more
In this article a new strategy is presented that can be used by neurophysicians, neurosurgeons, and orthopedic surgeons to predict patients' health after an operative procedure on the vertebral column just by analyzing the preoperative patient data. Usually, this is done based on the linguistic or heuristic variables related to patient's data, such as marital status, occupation, and so on.
Research Interests:
Eighty one percent of adolescents aged 11–17 years are inadequately physically active worldwide. Physical activity (PA) recommendations for high school children have not been studied previously in schools in the Qatar region. The... more
Eighty one percent of adolescents aged 11–17 years are inadequately physically active worldwide. Physical activity (PA) recommendations for high school children have not been studied previously in schools in the Qatar region. The objectives of the study were: (i) to assess the level of compliance of the recommended PA and to assess if there are any gender differences; and (ii) to analyze the recommended step count compliance during school and non-school days. An observational cross-sectional study was conducted. Twenty-nine children (12 boys and 17 girls) aged 13–17 years (15.24 ± 1.46) took part in this study. Participants wore Fitbit Charge 5 wrist bands for three weeks to collect various digital biomarkers including moderate-to-vigorous physical activity (MVPA) and step counts (tracking during out-of-school time and school time). Based on this study, high school children in the two Qatar region schools did not meet the MVPA and steps/day recommendation by the established agencies...
Research Interests:
Predicting metro ridership is an essential requirement for efficient metro operation and management. The dependence of metro ridership on the land use densities entails a need for an accurate predictive model. To this end, the current... more
Predicting metro ridership is an essential requirement for efficient metro operation and management. The dependence of metro ridership on the land use densities entails a need for an accurate predictive model. To this end, the current study is aimed to develop a novel machine learning (ML) based model to predict the metro station ridership utilizing the land use densities near metro stations. The ridership data was obtained from Qatar Rail, and the land use data were obtained from the Ministry of Municipality and Environment in Qatar. The land use densities in the catchment area of 800 m around the metro stations have been considered in this study. The non-linear relationship between the metro ridership and land use densities has been captured through different ensemble ML models including random forests, extremely randomized trees, and gradient tree boosting. Results showed that the ML models, once meticulously optimized and trained are capable of producing an accurate prediction f...
Research Interests:
Research Interests:
Research Interests:
Background and Objective Age-related macular-degeneration (AMD) is one of the most common reasons for blindness in the world today. The most common treatment for wet AMD is the intra-vitreal injections for inhibiting... more
Background and Objective Age-related macular-degeneration (AMD) is one of the most common reasons for blindness in the world today. The most common treatment for wet AMD is the intra-vitreal injections for inhibiting Vascular-Endothelial-Derived-Growth-Factor (VEGF). This treatment usually involves multiple injections and thus multiple clinic visits which not only causes increased cost on national health services but also causes exposure to the hospital environment which is sometimes high risk considering current COVID crisis. The treatment, in spite of the above concerns, is usually effective. However, in some cases, either the medicine fails to produce the anticipated favorable outcome, resulting in waste of time, medication, efforts, and above all, psychological distress to the patients. Hence, early predictability of anatomical as well as functional effectiveness of the treatment appears to be a very desirable capability to have. Method A Machine Learning approach using Adaptive...
Research Interests:
Research Interests:
Research Interests:
Roaming complexity in terrains and unexpected environments poses significant difficulties in robotic exploration of an area. In a broader sense, robots have to face two common tasks during exploration namely walking on the drylands and... more
Roaming complexity in terrains and unexpected environments poses significant difficulties in robotic exploration of an area. In a broader sense, robots have to face two common tasks during exploration namely walking on the drylands and swimming through the water. This research aims to design and develop an amphibious robot, which incorporates webbed duck feet design to walk on different terrains, swim in water and tackle obstructions on its way. The designed robot is compact and easy-to-use and also has the abilities to work autonomously. Such a mechanism is implemented by designing a novel robotic webbed foot consisting of two hinged plates. Due to the design, the webbed feet are able to open and close with the help of water pressure. The Klann linkage has been used to convert rotational motion to walking and swimming as an animal’s gait. Because of its amphibian nature, the designed robot can be used for exploring the tight caves, closed spaces and moving on uneven challenging ter...
Research Interests: Engineering, Robotics, Computer Science, Propulsion, Simulation, and 5 moreGait, ROBOT, Terrain, Roaming, and Preprints
Research Interests:
Research Interests:
Avoidingheavy mathematics and lengthy programming details, Digital Image Processing: An Algorithmic Approach with MATLAB presents an easy methodology for learning the fundamentals of image processing. The book applies the algorithms using... more
Avoidingheavy mathematics and lengthy programming details, Digital Image Processing: An Algorithmic Approach with MATLAB presents an easy methodology for learning the fundamentals of image processing. The book applies the algorithms using MATLAB, without bogging down students with syntactical and debugging issues. A complete one-semester course on digital image processingOne chapter can typically be completed per week, with each chapter divided into three sections. The first section presents theoretical topics in a very simple and basic style with generic language and mathematics. The second section explains the theoretical concepts using flowcharts to streamline the concepts and to form a foundation for students to code in any programming language. The final section supplies MATLAB codes for reproducing the figures presented in the chapter. Programming-based exercises at the end of each chapter facilitate the learning of underlying concepts through practice. This textbook equips undergraduate students in computer engineering and science with an essential understanding of digital image processing. It will also help them comprehend more advanced topics and sophisticated mathematical material in later courses. A color insert is included in the text while various instructor resources are available on the authors website.
Research Interests:
Determining the grade of colon cancer from tissue slides is a routine part of the pathological analysis. In the case of colorectal adenocarcinoma (CRA), grading is partly determined by morphology and degree of formation of glandular... more
Determining the grade of colon cancer from tissue slides is a routine part of the pathological analysis. In the case of colorectal adenocarcinoma (CRA), grading is partly determined by morphology and degree of formation of glandular structures. Achieving consistency between pathologists is difficult due to the subjective nature of grading assessment. An objective grading using computer algorithms will be more consistent, and will be able to analyse images in more detail. In this paper, we measure the shape of glands with a novel metric that we call the Best Alignment Metric (BAM). We show a strong correlation between a novel measure of glandular shape and grade of the tumour. We used shape specific parameters to perform a two-class classification of images into normal or cancerous tissue and a three-class classification into normal, low grade cancer, and high grade cancer. The task of detecting gland boundaries, which is a prerequisite of shape-based analysis, was carried out using ...
Research Interests:
Research Interests:
The prediction of infarction volume after stroke onset depends on the shape of the growth dynamics of the infarction. To understand growth patterns that predict lesion volume changes, we studied currently available models described in... more
The prediction of infarction volume after stroke onset depends on the shape of the growth dynamics of the infarction. To understand growth patterns that predict lesion volume changes, we studied currently available models described in literature and compared the models with Adaptive Neuro-Fuzzy Inference System [ANFIS], a method previously unused in the prediction of infarction growth and infarction volume (IV). We included 67 patients with malignant middle cerebral artery [MMCA] stroke who underwent decompressive hemicraniectomy. All patients had at least three cranial CT scans prior to the surgery. The rate of growth and volume of infarction measured on the third CT was predicted with ANFIS without statistically significant difference compared to the ground truth [P = 0.489]. This was not possible with linear, logarithmic or exponential methods. ANFIS was able to predict infarction volume [IV3] over a wide range of volume [163.7-600 cm(3)] and time [22-110 hours]. The cross correl...
Research Interests:
Research Interests:
Research Interests: Mathematics, Computer Science, Artificial Intelligence, Computer Vision, Nondestructive Evaluation, and 11 moreBlind Source Separation, Nondestructive testing, Deconvolution, Image Restoration, Blind Deconvolution, Filtering, Ultrasonic Sensor, Image Sensors, Image Formation, State Space Model, and Symmetric Matrices
Research Interests:
In this paper, an algorithmic approach is presented that utilizes the system modeling approach to map the infrared (IR) pixel space to real optical space. This mapping has been performed using the parametric approach with Recursive... more
In this paper, an algorithmic approach is presented that utilizes the system modeling approach to map the infrared (IR) pixel space to real optical space. This mapping has been performed using the parametric approach with Recursive Weighted Least Squares (RWLS) estimation algorithm. Using the autoregressive (ARX) model structure, the image pixels are mapped from the IR pixel space into normal optical image space, thus enhancing the IR image for improved visual quality.
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests: Neurology, Neurosurgery, Fuzzy Logic, Forecasting, Medicine, and 15 moreHydrocephalus, Humans, Complications, Female, Male, Osteomyelitis, Clinical Sciences, Adult, Bone Substitutes, Craniofacial Surgery, Brain Neoplasms, Cranioplasty, Decompressive Craniectomy, Brain injuries, and Bone Transplantation
Research Interests:
In this paper, two algorithms have been presented to enhance the infrared (IR) images. Using the autoregressive moving average model structure and H(infinity) optimal bounds, the image pixels are mapped from the IR pixel space into normal... more
In this paper, two algorithms have been presented to enhance the infrared (IR) images. Using the autoregressive moving average model structure and H(infinity) optimal bounds, the image pixels are mapped from the IR pixel space into normal optical image space, thus enhancing the IR image for improved visual quality. Although H(infinity)-based system identification algorithms are very common now, they are not quite suitable for real-time applications owing to their complexity. However, many variants of such algorithms are possible that can overcome this constraint. Two such algorithms have been developed and implemented in this paper. Theoretical and algorithmic results show remarkable enhancement in the acquired images. This will help in enhancing the visual quality of IR images for surveillance applications.