Tanvir Mahmud - Academia.edu (original) (raw)

Papers by Tanvir Mahmud

2020 IEEE Region 10 Symposium (TENSYMP)

Classifying handwritten digits is one of the most trending topics of research in the study of the... more Classifying handwritten digits is one of the most trending topics of research in the study of the automated text recognition system. The problem is more challenging in the case of Bengali digits due to additional complexities arising from similarity among various digits along with a wide variety of styles of hand-writings. In this paper, an end-to-end deep convolutional neural network, named as DeepBanglaNet, is proposed to classify Bengali handwritten digits. The proposed network utilizes various state-of-the-art optimization algorithms for eliminating vanishing/exploding gradient problems while extracting the global features effectively required for proper recognition of handwritten digits. This results in a very efficient model providing state-of-the-art accuracy of 99.43% on the NumtaDB database and outperforms all other existing models in all traditional evaluation metrics.

2020 IEEE Region 10 Symposium (TENSYMP)

The use of photonic crystal fiber (PCF) in a wide variety of applications involving THz communica... more The use of photonic crystal fiber (PCF) in a wide variety of applications involving THz communications, sensing useful and harmful gas and liquids, imaging and spectroscopy is getting well established with the researches of last two decades. We are demonstrating a novel design of a suspended core PCF for sensing alcohol in this paper. The shape of the designed core of the fiber is elliptical with circular air pores incorporated into it, while the cladding of the fiber has been designed to be hexagonal in shape filled completely with air that results in high relative sensitivity along with good birefringence. Extensive simulations have been carried out using COMSOL design environment to evaluate the proposed architecture. Proposed architecture provides state-of-the-art relative sensitivity of 80.31% with birefringence of 0.016 at 0.65 THz, which outperforms all other published architectures.

2020 42nd Annual International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC)

The topic of automatic detection of sleep apnea which is a respiratory sleep disorder, affecting ... more The topic of automatic detection of sleep apnea which is a respiratory sleep disorder, affecting millions of patients worldwide, is continuously being explored by researchers. Electroencephalogram signal (EEG) represents a promising tool due to its direct correlation to neural activity and ease of extraction. Here, an innovative approach is proposed to automatically detect apnea by incorporating local variations of temporal features for identifying the global feature variations over a broader window. An EEG data frame is divided into smaller sub-frames to effectively extract local feature variation within one larger frame. A fully convolutional neural network (FCNN) is proposed that will take each sub-frame of a single frame individually to extract local features. Following that, a dense classifier consisting of a series of fully connected layers is trained to analyze all the local features extracted from subframes for classifying the entire frame as apnea/non-apnea. Finally, a unique post-processing technique is applied which significantly improves accuracy. Both the EEG frame length and post-processing parameters are varied to find optimal detection conditions. Large-scale experimentation is executed on publicly available data of patients with varying apnea-hypopnea indices for performance evaluation of the suggested method.

2019 7th International Conference on Robot Intelligence Technology and Applications (RiTA), 2019

The Electrocardiogram(ECG) is a wide spread used tool to monitor the health of a human heart. Det... more The Electrocardiogram(ECG) is a wide spread used tool to monitor the health of a human heart. Detecting any abnormalities of heart signal is the primary objective. Researchers have given a great attention to make this detection error- less and to detect the heart beats abnormality as quick as possible. In this paper, we proposed a method to detect heart beats abnormality efficiently. Our proposed structure is quite lightweight requiring less computational power and memory. Furthermore, to reduce class imbalance while increasing accuracy, we preprocessed our data and augmented the lower numbered classes with 6 different operations. For arrhythmia classification, we achieved average accuracy of 97.3%, 98.9% with F1 score of 97.21%, 99.2% & specificity of 99.3%, 98.95% for MIT BIH Arrhythmia database and PTB Diagnostic ECG database respectively, which is higher enough for a lightweight architecture like proposed one.

2020 IEEE 63rd International Midwest Symposium on Circuits and Systems (MWSCAS), 2020

Activity recognition from wearable sensors is a promising field of research with a wide variety o... more Activity recognition from wearable sensors is a promising field of research with a wide variety of applications to track human activity from distant positions. In this paper, a multi-stage long short term memory (LSTM) based deep neural network is proposed to integrate multimodal features from numerous sensors for activity recognition. In the first stage, for separately extracting effective temporal features from each sensor, an individual stack of LSTM layers are introduced on each sensor data. Afterward, extracted features from numerous sensors are aggregated maintaining their temporal dependency. Finally, for joint optimization of the aggregated multimodal features, a global feature optimizer network is proposed consisting of multiple LSTM layers followed by series of densely connected layers that extracts the global features through the fusion of multimodal features. Extensive experimentations on a publicly available dataset provide very satisfactory performance with an average F1 score of 83.9%.

Automatic lung lesions segmentation of chest CT scans is considered a pivotal stage towards accur... more Automatic lung lesions segmentation of chest CT scans is considered a pivotal stage towards accurate diagnosis and severity measurement of COVID-19. Traditional U-shaped encoder-decoder architecture and its variants suffer from diminutions of contextual information in pooling/upsampling operations with increased semantic gaps among encoded and decoded feature maps as well as instigate vanishing gradient problems for its sequential gradient propagation that result in sub-optimal performance. Moreover, operating with 3D CT-volume poses further limitations due to the exponential increase of computational complexity making the optimization difficult. In this paper, an automated COVID-19 lesion segmentation scheme is proposed utilizing a highly efficient neural network architecture, namely CovSegNet, to overcome these limitations. Additionally, a two-phase training scheme is introduced where a deeper 2D-network is employed for generating ROI-enhanced CT-volume followed by a shallower 3D-...

2020 IEEE Region 10 Symposium (TENSYMP), 2020

Sleep apnea is a sleep disorder that millions of people all over the world are affected with. Unt... more Sleep apnea is a sleep disorder that millions of people all over the world are affected with. Untreated Sleep apnea can lead to various complex health issues including death. The detection of apnea events has been a pressing topic of research in the recent years. Several signals like polysomnography (PSG), electrocardiogram (ECG), electroencephalogram (EEG) are used to detect sleep apnea. In this paper, a novel approach has been proposed using EEG signal. The decomposed EEG signal is fed into a Long Short Term Memory (LSTM) model to explore the sequence of the signals. The output is then used for a Deep Neural Network (DNN) to correctly detect apnea frames. Lastly all the predictions are post-processed to get the final result. This scheme has the potential to be used in hospitals for continuous detection of sleep apnea event.

IEEE Sensors Journal, 2021

Deep neural network is an effective choice to automatically recognize human actions utilizing dat... more Deep neural network is an effective choice to automatically recognize human actions utilizing data from various wearable sensors. These networks automate the process of feature extraction relying completely on data. However, various noises in time series data with complex intermodal relationships among sensors make this process more complicated. In this paper, we have proposed a novel multi-stage training approach that increases diversity in this feature extraction process to make accurate recognition of actions by combining varieties of features extracted from diverse perspectives. Initially, instead of using single type of transformation, numerous transformations are employed on time series data to obtain variegated representations of the features encoded in raw data. An efficient deep CNN architecture is proposed that can be individually trained to extract features from different transformed spaces. Later, these CNN feature extractors are merged into an optimal architecture finely tuned for optimizing diversified extracted features through a combined training stage or multiple sequential training stages. This approach offers the opportunity to explore the encoded features in raw sensor data utilizing multifarious observation windows with immense scope for efficient selection of features for final convergence. Extensive experimentations have been carried out in three publicly available datasets that provide outstanding performance consistently with average five-fold cross-validation accuracy of 99.29% on UCI HAR database, 99.02% on USC HAR database, and 97.21% on SKODA database outperforming other state-of-the-art approaches.

ArXiv, 2020

Malaria, a life-threatening disease, infects millions of people every year throughout the world d... more Malaria, a life-threatening disease, infects millions of people every year throughout the world demanding faster diagnosis for proper treatment before any damages occur. In this paper, an end-to-end deep learning-based approach is proposed for faster diagnosis of malaria from thin blood smear images by making efficient optimizations of features extracted from diversified receptive fields. Firstly, an efficient, highly scalable deep neural network, named as DilationNet, is proposed that incorporates features from a large spectrum by varying dilation rates of convolutions to extract features from different receptive areas. Next, the raw images are resampled to various resolutions to introduce variations in the receptive fields that are used for independently optimizing different forms of DilationNet scaled for different resolutions of images. Afterward, a feature fusion scheme is introduced with the proposed DeepFusionNet architecture for jointly optimizing the feature space of these ...

2020 IEEE REGION 10 CONFERENCE (TENCON), 2020

Since the onset of COVID-19, radiographic image analysis coupled with artificial intelligence (AI... more Since the onset of COVID-19, radiographic image analysis coupled with artificial intelligence (AI) has become popular due to insufficient RT-PCR test kits. In this paper, an automated AI-assisted COVID-19 diagnosis scheme is proposed utilizing the ensembling approach of multiple convolutional neural networks (CNNs). Two different strategies have been carried out for ensembling: A feature level fusionbased ensembling method and a decision level ensembling method. Several traditional CNN architectures are tested and finally in the ensembling operation, MobileNet, InceptionV3, DenseNet201, DenseNet121 and Xception are used. To handle the computational complexity of multiple networks, transfer learning strategy is incorporated through ImageNet pre-trained weight initialization. For feature-level ensembling scheme, global averages of the convolutional feature maps generated from multiple networks are aggregated and undergo through fully connected layers for combined optimization. Additionally, for decision level ensembling scheme, final prediction generated from multiple networks are converged into a single prediction by utilizing the maximum voting criterion. Both strategies perform better than any individual network. Outstanding performances have been achieved through extensive experimentation on a public database with 96% accuracy on 3-class (COVID-19/normal/pneumonia) diagnosis and 89.21% on 4class (COVID-19/normal/viral pneumonia/bacterial pneumonia) diagnosis.

2020 IEEE REGION 10 CONFERENCE (TENCON), 2020

Radiology examination of chest radiography or chest X-ray (CXR), is currently performed manually ... more Radiology examination of chest radiography or chest X-ray (CXR), is currently performed manually by radiologists. With the onset of the COVID-19 pandemic, there is now a need to automate this process which is currently one of the key methods of primary detection of the SARS-Cov-2 virus. This will lead to shorter diagnosis time and less human error. In this study, we try to perform three-class image classification on a dataset of chest X-rays of confirmed COVID-19 patients(408 images), confirmed pneumonia patients(4273 images), and chest X-rays of healthy people(1590 images). In total the dataset consists of 6271 people. We aim to use a Convolutional Neural Network(CNN) and transfer learning to perform this image classification task. Our model is based on a pre-trained InceptionV3 network with weights trained on the ImageNet dataset. We fine-tune the layers of the Inception network to train it to our specific task. We try fine-tuning the network to different extents by freezing a different number of layers and then comparing accuracy for each variation of the network. To evaluate the performance of our network we use several metrics which include Classification accuracy, Precision, Sensitivity, and Specificity. Our proposed method achieves an accuracy of 96.33% on a 3-class classification task (Normal, COVID-19, Pneumonia) and an accuracy of 99.39% on a 2-class (COVID and Non-COVID) classification task.

2020 11th International Conference on Electrical and Computer Engineering (ICECE), 2020

Sleep apnea, a common sleep disorder, has been affecting millions of people all over the world. F... more Sleep apnea, a common sleep disorder, has been affecting millions of people all over the world. For automatic detection of sleep apnea from various bio-signals, the Electroencephalogram (EEG) signal is getting more attention because of its physiological interpretation with this disease. In this paper, a patient independent sub-frame based approach for the automatic detection of apnea frames using only EEG signal is proposed. Here instead of directly using a whole frame of EEG data, spectro-temporal subframes are used that are obtained by first extracting frequency band limited signals and then dividing each of them into smaller subframes. Next the extracted subframes are fed into the proposed local convolutional neural network (CNN) blocks. The local features thus produced are then processed using the proposed global CNN block to obtain global features. These features are optimized using deep neural network classifier. The method is evaluated on multiple patients taken from a publicly available database. From extensive analysis it is found that the proposed method offers consistently significant performance in terms of accuracy, sensitivity and specificity. The proposed scheme has the potential to be used for the better detection of sleep apnea in real life application.

IEEE Transactions on Industrial Informatics, 2021

Rapid and precise diagnosis of COVID-19 is one of the major challenges faced by the global commun... more Rapid and precise diagnosis of COVID-19 is one of the major challenges faced by the global community to control the spread of this overgrowing pandemic. In this article, a hybrid neural network is proposed, named Cov-TANet, to provide an end-to-end clinical diagnostic tool for early diagnosis, lesion segmentation, and severity prediction of COVID-19 utilizing chest computer tomography (CT) scans. A multiphase optimization strategy is introduced for solving the challenges of complicated diagnosis at a very early stage of infection, where an efficient lesion segmentation network is optimized initially, which is later integrated into a joint optimization framework for the diagnosis and severity prediction tasks providing feature enhancement of the infected regions. Moreover, for overcoming the challenges with diffused, blurred, and varying shaped edges of COVID lesions with novel and diverse characteristics, a novel segmentation network is introduced, namely trilevel attention-based segmentation network. This network has significantly reduced semantic gaps in subsequent encoding-decoding stages, with immense parallelization of multiscale features for faster convergence providing considerable performance improvement over traditional networks. Furthermore, a novel tri-level attention mechanism has been introduced, which is repeatedly utilized over the network, combining channel, spatial, and pixel attention

Computers in Biology and Medicine, 2020

With the recent outbreak of COVID-19, fast diagnostic testing has become one of the major challen... more With the recent outbreak of COVID-19, fast diagnostic testing has become one of the major challenges due to the critical shortage of test kit. Pneumonia, a major effect of COVID-19, needs to be urgently diagnosed along with its underlying reasons. In this paper, deep learning aided automated COVID-19 and other pneumonia detection schemes are proposed utilizing a small amount of COVID-19 chest X-rays. A deep convolutional neural network (CNN) based architecture, named as CovXNet, is proposed that utilizes depthwise convolution with varying dilation rates for efficiently extracting diversified features from chest X-rays. Since the chest Xray images corresponding to COVID-19 caused pneumonia and other traditional pneumonias have significant similarities, at first, a large number of chest X-rays corresponding to normal and (viral/bacterial) pneumonia patients are used to train the proposed CovXNet. Learning of this initial training phase is transferred with some additional fine-tuning layers that are further trained with a smaller number of chest X-rays corresponding to COVID-19 and other pneumonia patients. In the proposed method, different forms of CovXNets are designed and trained with X-ray images of various resolutions and for further optimization of their predictions, a stacking algorithm is employed. Finally, a gradient-based discriminative localization is integrated to distinguish the abnormal regions of X-ray images referring to different types of pneumonia. Extensive experimentations using two different datasets provide very satisfactory detection performance with accuracy of 97.4% for COVID/Normal, 96.9% for COVID/Viral pneumonia, 94.7% for COVID/Bacterial pneumonia, and 90.2% for multiclass COVID/normal/Viral/Bacterial pneumonias. Hence, the proposed schemes can serve as an efficient tool in the current state of COVID-19 pandemic. All the architectures are made publicly available at: https: //github.com/Perceptron21/CovXNet.

Journal of interpersonal violence, 2018

We examined female participation in household decision making and its association with the justif... more We examined female participation in household decision making and its association with the justification of wife beating in Bangladesh. We used nationally representative data from the 2014 Bangladesh Demographic and Health Survey. Our sample consisted of currently married women of age 15 to 49 years ( n = 16,463). Chi-square tests and multilevel logistic regression models were performed. Approximately 84% of women in the survey were participants in at least one household decision, and 72% reported that wife beating is not justified in any circumstance. Women who reported their participation in at least one type of household decision less frequently reported that wife beating could be justified than those who did not participate in any household decisions (adjusted odds ratio = 1.49; 95% confidence interval = [1.25, 1.78]). In addition to participation in household decision making, other factors including age at first marriage, females' and their husbands' education, religion...

2019 IEEE International Conference on Robotics, Automation, Artificial-intelligence and Internet-of-Things (RAAICON)

In high-rise buildings with a large number of passersby, banks of elevators working in parallel a... more In high-rise buildings with a large number of passersby, banks of elevators working in parallel are required to serve the populace efficiently and quickly. For this purpose, a generalized control mechanism has been developed for any number of elevators such that each call will be served by the elevator deemed to be the most energy-efficient. The control mechanism used here is derived from the ground up using simple calculations so that it will be computationally cheap, fast, and implementable with the most basic microcontroller. First, a single elevator control mechanism was developed; next, the idea was extended toward the common control mechanism for multiple elevators such that every common call would be referred to the elevator that is best suited for it. The issue of controlling the system using novel mechanisms and simplest possible calculations was prioritized.

Computers in Biology and Medicine

2020 IEEE Region 10 Symposium (TENSYMP)

Classifying handwritten digits is one of the most trending topics of research in the study of the... more Classifying handwritten digits is one of the most trending topics of research in the study of the automated text recognition system. The problem is more challenging in the case of Bengali digits due to additional complexities arising from similarity among various digits along with a wide variety of styles of hand-writings. In this paper, an end-to-end deep convolutional neural network, named as DeepBanglaNet, is proposed to classify Bengali handwritten digits. The proposed network utilizes various state-of-the-art optimization algorithms for eliminating vanishing/exploding gradient problems while extracting the global features effectively required for proper recognition of handwritten digits. This results in a very efficient model providing state-of-the-art accuracy of 99.43% on the NumtaDB database and outperforms all other existing models in all traditional evaluation metrics.

2020 IEEE Region 10 Symposium (TENSYMP)

The use of photonic crystal fiber (PCF) in a wide variety of applications involving THz communica... more The use of photonic crystal fiber (PCF) in a wide variety of applications involving THz communications, sensing useful and harmful gas and liquids, imaging and spectroscopy is getting well established with the researches of last two decades. We are demonstrating a novel design of a suspended core PCF for sensing alcohol in this paper. The shape of the designed core of the fiber is elliptical with circular air pores incorporated into it, while the cladding of the fiber has been designed to be hexagonal in shape filled completely with air that results in high relative sensitivity along with good birefringence. Extensive simulations have been carried out using COMSOL design environment to evaluate the proposed architecture. Proposed architecture provides state-of-the-art relative sensitivity of 80.31% with birefringence of 0.016 at 0.65 THz, which outperforms all other published architectures.

2020 42nd Annual International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC)

The topic of automatic detection of sleep apnea which is a respiratory sleep disorder, affecting ... more The topic of automatic detection of sleep apnea which is a respiratory sleep disorder, affecting millions of patients worldwide, is continuously being explored by researchers. Electroencephalogram signal (EEG) represents a promising tool due to its direct correlation to neural activity and ease of extraction. Here, an innovative approach is proposed to automatically detect apnea by incorporating local variations of temporal features for identifying the global feature variations over a broader window. An EEG data frame is divided into smaller sub-frames to effectively extract local feature variation within one larger frame. A fully convolutional neural network (FCNN) is proposed that will take each sub-frame of a single frame individually to extract local features. Following that, a dense classifier consisting of a series of fully connected layers is trained to analyze all the local features extracted from subframes for classifying the entire frame as apnea/non-apnea. Finally, a unique post-processing technique is applied which significantly improves accuracy. Both the EEG frame length and post-processing parameters are varied to find optimal detection conditions. Large-scale experimentation is executed on publicly available data of patients with varying apnea-hypopnea indices for performance evaluation of the suggested method.

2019 7th International Conference on Robot Intelligence Technology and Applications (RiTA), 2019

The Electrocardiogram(ECG) is a wide spread used tool to monitor the health of a human heart. Det... more The Electrocardiogram(ECG) is a wide spread used tool to monitor the health of a human heart. Detecting any abnormalities of heart signal is the primary objective. Researchers have given a great attention to make this detection error- less and to detect the heart beats abnormality as quick as possible. In this paper, we proposed a method to detect heart beats abnormality efficiently. Our proposed structure is quite lightweight requiring less computational power and memory. Furthermore, to reduce class imbalance while increasing accuracy, we preprocessed our data and augmented the lower numbered classes with 6 different operations. For arrhythmia classification, we achieved average accuracy of 97.3%, 98.9% with F1 score of 97.21%, 99.2% & specificity of 99.3%, 98.95% for MIT BIH Arrhythmia database and PTB Diagnostic ECG database respectively, which is higher enough for a lightweight architecture like proposed one.

2020 IEEE 63rd International Midwest Symposium on Circuits and Systems (MWSCAS), 2020

Activity recognition from wearable sensors is a promising field of research with a wide variety o... more Activity recognition from wearable sensors is a promising field of research with a wide variety of applications to track human activity from distant positions. In this paper, a multi-stage long short term memory (LSTM) based deep neural network is proposed to integrate multimodal features from numerous sensors for activity recognition. In the first stage, for separately extracting effective temporal features from each sensor, an individual stack of LSTM layers are introduced on each sensor data. Afterward, extracted features from numerous sensors are aggregated maintaining their temporal dependency. Finally, for joint optimization of the aggregated multimodal features, a global feature optimizer network is proposed consisting of multiple LSTM layers followed by series of densely connected layers that extracts the global features through the fusion of multimodal features. Extensive experimentations on a publicly available dataset provide very satisfactory performance with an average F1 score of 83.9%.

Automatic lung lesions segmentation of chest CT scans is considered a pivotal stage towards accur... more Automatic lung lesions segmentation of chest CT scans is considered a pivotal stage towards accurate diagnosis and severity measurement of COVID-19. Traditional U-shaped encoder-decoder architecture and its variants suffer from diminutions of contextual information in pooling/upsampling operations with increased semantic gaps among encoded and decoded feature maps as well as instigate vanishing gradient problems for its sequential gradient propagation that result in sub-optimal performance. Moreover, operating with 3D CT-volume poses further limitations due to the exponential increase of computational complexity making the optimization difficult. In this paper, an automated COVID-19 lesion segmentation scheme is proposed utilizing a highly efficient neural network architecture, namely CovSegNet, to overcome these limitations. Additionally, a two-phase training scheme is introduced where a deeper 2D-network is employed for generating ROI-enhanced CT-volume followed by a shallower 3D-...

2020 IEEE Region 10 Symposium (TENSYMP), 2020

Sleep apnea is a sleep disorder that millions of people all over the world are affected with. Unt... more Sleep apnea is a sleep disorder that millions of people all over the world are affected with. Untreated Sleep apnea can lead to various complex health issues including death. The detection of apnea events has been a pressing topic of research in the recent years. Several signals like polysomnography (PSG), electrocardiogram (ECG), electroencephalogram (EEG) are used to detect sleep apnea. In this paper, a novel approach has been proposed using EEG signal. The decomposed EEG signal is fed into a Long Short Term Memory (LSTM) model to explore the sequence of the signals. The output is then used for a Deep Neural Network (DNN) to correctly detect apnea frames. Lastly all the predictions are post-processed to get the final result. This scheme has the potential to be used in hospitals for continuous detection of sleep apnea event.

IEEE Sensors Journal, 2021

Deep neural network is an effective choice to automatically recognize human actions utilizing dat... more Deep neural network is an effective choice to automatically recognize human actions utilizing data from various wearable sensors. These networks automate the process of feature extraction relying completely on data. However, various noises in time series data with complex intermodal relationships among sensors make this process more complicated. In this paper, we have proposed a novel multi-stage training approach that increases diversity in this feature extraction process to make accurate recognition of actions by combining varieties of features extracted from diverse perspectives. Initially, instead of using single type of transformation, numerous transformations are employed on time series data to obtain variegated representations of the features encoded in raw data. An efficient deep CNN architecture is proposed that can be individually trained to extract features from different transformed spaces. Later, these CNN feature extractors are merged into an optimal architecture finely tuned for optimizing diversified extracted features through a combined training stage or multiple sequential training stages. This approach offers the opportunity to explore the encoded features in raw sensor data utilizing multifarious observation windows with immense scope for efficient selection of features for final convergence. Extensive experimentations have been carried out in three publicly available datasets that provide outstanding performance consistently with average five-fold cross-validation accuracy of 99.29% on UCI HAR database, 99.02% on USC HAR database, and 97.21% on SKODA database outperforming other state-of-the-art approaches.

ArXiv, 2020

Malaria, a life-threatening disease, infects millions of people every year throughout the world d... more Malaria, a life-threatening disease, infects millions of people every year throughout the world demanding faster diagnosis for proper treatment before any damages occur. In this paper, an end-to-end deep learning-based approach is proposed for faster diagnosis of malaria from thin blood smear images by making efficient optimizations of features extracted from diversified receptive fields. Firstly, an efficient, highly scalable deep neural network, named as DilationNet, is proposed that incorporates features from a large spectrum by varying dilation rates of convolutions to extract features from different receptive areas. Next, the raw images are resampled to various resolutions to introduce variations in the receptive fields that are used for independently optimizing different forms of DilationNet scaled for different resolutions of images. Afterward, a feature fusion scheme is introduced with the proposed DeepFusionNet architecture for jointly optimizing the feature space of these ...

2020 IEEE REGION 10 CONFERENCE (TENCON), 2020

Since the onset of COVID-19, radiographic image analysis coupled with artificial intelligence (AI... more Since the onset of COVID-19, radiographic image analysis coupled with artificial intelligence (AI) has become popular due to insufficient RT-PCR test kits. In this paper, an automated AI-assisted COVID-19 diagnosis scheme is proposed utilizing the ensembling approach of multiple convolutional neural networks (CNNs). Two different strategies have been carried out for ensembling: A feature level fusionbased ensembling method and a decision level ensembling method. Several traditional CNN architectures are tested and finally in the ensembling operation, MobileNet, InceptionV3, DenseNet201, DenseNet121 and Xception are used. To handle the computational complexity of multiple networks, transfer learning strategy is incorporated through ImageNet pre-trained weight initialization. For feature-level ensembling scheme, global averages of the convolutional feature maps generated from multiple networks are aggregated and undergo through fully connected layers for combined optimization. Additionally, for decision level ensembling scheme, final prediction generated from multiple networks are converged into a single prediction by utilizing the maximum voting criterion. Both strategies perform better than any individual network. Outstanding performances have been achieved through extensive experimentation on a public database with 96% accuracy on 3-class (COVID-19/normal/pneumonia) diagnosis and 89.21% on 4class (COVID-19/normal/viral pneumonia/bacterial pneumonia) diagnosis.

2020 IEEE REGION 10 CONFERENCE (TENCON), 2020

Radiology examination of chest radiography or chest X-ray (CXR), is currently performed manually ... more Radiology examination of chest radiography or chest X-ray (CXR), is currently performed manually by radiologists. With the onset of the COVID-19 pandemic, there is now a need to automate this process which is currently one of the key methods of primary detection of the SARS-Cov-2 virus. This will lead to shorter diagnosis time and less human error. In this study, we try to perform three-class image classification on a dataset of chest X-rays of confirmed COVID-19 patients(408 images), confirmed pneumonia patients(4273 images), and chest X-rays of healthy people(1590 images). In total the dataset consists of 6271 people. We aim to use a Convolutional Neural Network(CNN) and transfer learning to perform this image classification task. Our model is based on a pre-trained InceptionV3 network with weights trained on the ImageNet dataset. We fine-tune the layers of the Inception network to train it to our specific task. We try fine-tuning the network to different extents by freezing a different number of layers and then comparing accuracy for each variation of the network. To evaluate the performance of our network we use several metrics which include Classification accuracy, Precision, Sensitivity, and Specificity. Our proposed method achieves an accuracy of 96.33% on a 3-class classification task (Normal, COVID-19, Pneumonia) and an accuracy of 99.39% on a 2-class (COVID and Non-COVID) classification task.

2020 11th International Conference on Electrical and Computer Engineering (ICECE), 2020

Sleep apnea, a common sleep disorder, has been affecting millions of people all over the world. F... more Sleep apnea, a common sleep disorder, has been affecting millions of people all over the world. For automatic detection of sleep apnea from various bio-signals, the Electroencephalogram (EEG) signal is getting more attention because of its physiological interpretation with this disease. In this paper, a patient independent sub-frame based approach for the automatic detection of apnea frames using only EEG signal is proposed. Here instead of directly using a whole frame of EEG data, spectro-temporal subframes are used that are obtained by first extracting frequency band limited signals and then dividing each of them into smaller subframes. Next the extracted subframes are fed into the proposed local convolutional neural network (CNN) blocks. The local features thus produced are then processed using the proposed global CNN block to obtain global features. These features are optimized using deep neural network classifier. The method is evaluated on multiple patients taken from a publicly available database. From extensive analysis it is found that the proposed method offers consistently significant performance in terms of accuracy, sensitivity and specificity. The proposed scheme has the potential to be used for the better detection of sleep apnea in real life application.

IEEE Transactions on Industrial Informatics, 2021

Rapid and precise diagnosis of COVID-19 is one of the major challenges faced by the global commun... more Rapid and precise diagnosis of COVID-19 is one of the major challenges faced by the global community to control the spread of this overgrowing pandemic. In this article, a hybrid neural network is proposed, named Cov-TANet, to provide an end-to-end clinical diagnostic tool for early diagnosis, lesion segmentation, and severity prediction of COVID-19 utilizing chest computer tomography (CT) scans. A multiphase optimization strategy is introduced for solving the challenges of complicated diagnosis at a very early stage of infection, where an efficient lesion segmentation network is optimized initially, which is later integrated into a joint optimization framework for the diagnosis and severity prediction tasks providing feature enhancement of the infected regions. Moreover, for overcoming the challenges with diffused, blurred, and varying shaped edges of COVID lesions with novel and diverse characteristics, a novel segmentation network is introduced, namely trilevel attention-based segmentation network. This network has significantly reduced semantic gaps in subsequent encoding-decoding stages, with immense parallelization of multiscale features for faster convergence providing considerable performance improvement over traditional networks. Furthermore, a novel tri-level attention mechanism has been introduced, which is repeatedly utilized over the network, combining channel, spatial, and pixel attention

Computers in Biology and Medicine, 2020

With the recent outbreak of COVID-19, fast diagnostic testing has become one of the major challen... more With the recent outbreak of COVID-19, fast diagnostic testing has become one of the major challenges due to the critical shortage of test kit. Pneumonia, a major effect of COVID-19, needs to be urgently diagnosed along with its underlying reasons. In this paper, deep learning aided automated COVID-19 and other pneumonia detection schemes are proposed utilizing a small amount of COVID-19 chest X-rays. A deep convolutional neural network (CNN) based architecture, named as CovXNet, is proposed that utilizes depthwise convolution with varying dilation rates for efficiently extracting diversified features from chest X-rays. Since the chest Xray images corresponding to COVID-19 caused pneumonia and other traditional pneumonias have significant similarities, at first, a large number of chest X-rays corresponding to normal and (viral/bacterial) pneumonia patients are used to train the proposed CovXNet. Learning of this initial training phase is transferred with some additional fine-tuning layers that are further trained with a smaller number of chest X-rays corresponding to COVID-19 and other pneumonia patients. In the proposed method, different forms of CovXNets are designed and trained with X-ray images of various resolutions and for further optimization of their predictions, a stacking algorithm is employed. Finally, a gradient-based discriminative localization is integrated to distinguish the abnormal regions of X-ray images referring to different types of pneumonia. Extensive experimentations using two different datasets provide very satisfactory detection performance with accuracy of 97.4% for COVID/Normal, 96.9% for COVID/Viral pneumonia, 94.7% for COVID/Bacterial pneumonia, and 90.2% for multiclass COVID/normal/Viral/Bacterial pneumonias. Hence, the proposed schemes can serve as an efficient tool in the current state of COVID-19 pandemic. All the architectures are made publicly available at: https: //github.com/Perceptron21/CovXNet.

Journal of interpersonal violence, 2018

We examined female participation in household decision making and its association with the justif... more We examined female participation in household decision making and its association with the justification of wife beating in Bangladesh. We used nationally representative data from the 2014 Bangladesh Demographic and Health Survey. Our sample consisted of currently married women of age 15 to 49 years ( n = 16,463). Chi-square tests and multilevel logistic regression models were performed. Approximately 84% of women in the survey were participants in at least one household decision, and 72% reported that wife beating is not justified in any circumstance. Women who reported their participation in at least one type of household decision less frequently reported that wife beating could be justified than those who did not participate in any household decisions (adjusted odds ratio = 1.49; 95% confidence interval = [1.25, 1.78]). In addition to participation in household decision making, other factors including age at first marriage, females' and their husbands' education, religion...

2019 IEEE International Conference on Robotics, Automation, Artificial-intelligence and Internet-of-Things (RAAICON)

In high-rise buildings with a large number of passersby, banks of elevators working in parallel a... more In high-rise buildings with a large number of passersby, banks of elevators working in parallel are required to serve the populace efficiently and quickly. For this purpose, a generalized control mechanism has been developed for any number of elevators such that each call will be served by the elevator deemed to be the most energy-efficient. The control mechanism used here is derived from the ground up using simple calculations so that it will be computationally cheap, fast, and implementable with the most basic microcontroller. First, a single elevator control mechanism was developed; next, the idea was extended toward the common control mechanism for multiple elevators such that every common call would be referred to the elevator that is best suited for it. The issue of controlling the system using novel mechanisms and simplest possible calculations was prioritized.

Computers in Biology and Medicine