Keshav Thapa
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Keshav Thapa has a total of 22 co-authors. Here are the top co-authors by number of shared publications:
- Sung-Hyun Yang
Department of Electronic Engineering, Kwangwoon University, Seoul 01897, Korea
7 shared publications - Zubaer Md. Abdullah Al
Electronic Engineering Department, Kwangwoon University, Seoul 139-701, Korea
2 shared publications - Barsha Lamichhane
Department of Electronics Engineering, Kwangwoon University, Seoul 139-701, Korea
1 shared publication - Barsha Lamichhane
Department of Electronic Engineering, Kwangwoon University, 20 Kwangwoon-ro, Nowon-gu, Seoul 01897, Republic of Korea
1 shared publication - D.H. Lee
Department of Electronic Engineering, Kwangwoon University, Seoul 139-701, Republic of Korea
1 shared publication - Dong-Gwon Baek
Department of Electronic Engineering, Kwangwoon University, Seoul 01897, Korea
1 shared publication
Latest Publications
- Open Access
- 1 January 2024
- Published by Wiley in International Journal of Energy Research
- Vol. 2024 (1)
Recent Progress in Wearable Triboelectric Nanogenerator for Advanced Health Monitoring and Rehabilitation
Triboelectric nanogenerators (TENGs) have become a vital technology in physical health monitoring and rehabilitation applications, enabling continuous, personalized, and convenient monitoring, facilitating early detection and prevention, and offering valuable data‐driven insights. On‐body wearable triboelectric sensors enable real‐time tracking of vital health parameters, promoting accurate and personalized healthcare interventions. In addition to being self‐powered, these sensors facilitate early detection and prevention by capturing early warning signs of potential health issues, leading to timely interventions. Furthermore, they play a vital role in rehabilitation monitoring and feedback, helping to assess progress during the recovery process. This continuous monitoring, along with artificial intelligence, provides data‐driven insights into patterns, trends, and correlations, facilitating evidence‐based healthcare decision‐making. In this review, we provide an overview of the recent wearable TENG developments for healthcare applications. We discuss how TENGs work and explore a wide range of self‐powered health monitoring and rehabilitation applications. We then summarize recent advancements in healthcare applications, material selection, and unique features to help design highly efficient, accurate, and practical bioapplicable technology. Further, we discuss the challenges and prospects of the TENGs technology for healthcare applications, including miniaturization, moisture resistance, device maintenance, data reliability, and high outpower requirements.
- 1 January 2024
- Published by Wiley in International Journal of Energy Research
- Vol. 2024 (1)
Open AccessRecent Progress in Wearable Triboelectric Nanogenerator for Advanced Health Monitoring and Rehabilitation
Triboelectric nanogenerators (TENGs) have become a vital technology in physical health monitoring and rehabilitation applications, enabling continuous, personalized, and convenient monitoring, facilitating early detection and prevention, and offering valuable data‐driven insights. On‐body wearable triboelectric sensors enable real‐time tracking of vital health parameters, promoting accurate and personalized healthcare interventions. In addition to being self‐powered, these sensors facilitate early detection and prevention by capturing early warning signs of potential health issues, leading to timely interventions. Furthermore, they play a vital role in rehabilitation monitoring and feedback, helping to assess progress during the recovery process. This continuous monitoring, along with artificial intelligence, provides data‐driven insights into patterns, trends, and correlations, facilitating evidence‐based healthcare decision‐making. In this review, we provide an overview of the recent wearable TENG developments for healthcare applications. We discuss how TENGs work and explore a wide range of self‐powered health monitoring and rehabilitation applications. We then summarize recent advancements in healthcare applications, material selection, and unique features to help design highly efficient, accurate, and practical bioapplicable technology. Further, we discuss the challenges and prospects of the TENGs technology for healthcare applications, including miniaturization, moisture resistance, device maintenance, data reliability, and high outpower requirements.
- Open Access
Semi-Supervised Adversarial Auto-Encoder to Expedite Human Activity Recognition
The study of human activity recognition concentrates on classifying human activities and the inference of human behavior using modern sensing technology. However, the issue of domain adaptation for inertial sensing-based human activity recognition (HAR) is still burdensome. The existing requirement of labeled training data for adapting such classifiers to every new person, device, or on-body location is a significant barrier to the widespread adoption of HAR-based applications, making this a challenge of high practical importance. We propose the semi-supervised HAR method to improve reconstruction and generation. It executes proper adaptation with unlabeled data without changes to a pre-trained HAR classifier. Our approach decouples VAE with adversarial learning to ensure robust classifier operation, without newly labeled training data, under changes to the individual activity and the on-body sensor position. Our proposed framework shows the empirical results using the publicly available benchmark dataset compared to state-of-art baselines, achieving competitive improvement for handling new and unlabeled activity. The result demonstrates SAA has achieved a 5% improvement in classification score compared to the existing HAR platform.
Open AccessSemi-Supervised Adversarial Auto-Encoder to Expedite Human Activity Recognition
The study of human activity recognition concentrates on classifying human activities and the inference of human behavior using modern sensing technology. However, the issue of domain adaptation for inertial sensing-based human activity recognition (HAR) is still burdensome. The existing requirement of labeled training data for adapting such classifiers to every new person, device, or on-body location is a significant barrier to the widespread adoption of HAR-based applications, making this a challenge of high practical importance. We propose the semi-supervised HAR method to improve reconstruction and generation. It executes proper adaptation with unlabeled data without changes to a pre-trained HAR classifier. Our approach decouples VAE with adversarial learning to ensure robust classifier operation, without newly labeled training data, under changes to the individual activity and the on-body sensor position. Our proposed framework shows the empirical results using the publicly available benchmark dataset compared to state-of-art baselines, achieving competitive improvement for handling new and unlabeled activity. The result demonstrates SAA has achieved a 5% improvement in classification score compared to the existing HAR platform.
- Open Access
Detection of Image Level Forgery with Various Constraints Using DFDC Full and Sample Datasets
The emergence of advanced machine learning or deep learning techniques such as autoencoders and generative adversarial networks, can generate images known as deepfakes, which astonishingly resemble the realistic images. These deepfake images are hard to distinguish from the real images and are being used unethically against famous personalities such as politicians, celebrities, and social workers. Hence, we propose a method to detect these deepfake images using a light weighted convolutional neural network (CNN). Our research is conducted with Deep Fake Detection Challenge (DFDC) full and sample datasets, where we compare the performance of our proposed model with various state-of-the-art pretrained models such as VGG-19, Xception and Inception-ResNet-v2. Furthermore, we perform the experiments with various resolutions maintaining 1:1 and 9:16 aspect ratios, which have not been explored for DFDC datasets by any other groups to date. Thus, the proposed model can flexibly accommodate various resolutions and aspect ratios, without being constrained to a specific resolution or aspect ratio for any type of image classification problem. While most of the reported research is limited to sample or preview DFDC datasets only, we have also attempted the testing on full DFDC datasets and presented the results. Contemplating the fact that the detailed results and resource analysis for various scenarios are provided in this research, the proposed deepfake detection method is anticipated to pave new avenues for deepfake detection research, that engages with DFDC datasets.
Open AccessDetection of Image Level Forgery with Various Constraints Using DFDC Full and Sample Datasets
The emergence of advanced machine learning or deep learning techniques such as autoencoders and generative adversarial networks, can generate images known as deepfakes, which astonishingly resemble the realistic images. These deepfake images are hard to distinguish from the real images and are being used unethically against famous personalities such as politicians, celebrities, and social workers. Hence, we propose a method to detect these deepfake images using a light weighted convolutional neural network (CNN). Our research is conducted with Deep Fake Detection Challenge (DFDC) full and sample datasets, where we compare the performance of our proposed model with various state-of-the-art pretrained models such as VGG-19, Xception and Inception-ResNet-v2. Furthermore, we perform the experiments with various resolutions maintaining 1:1 and 9:16 aspect ratios, which have not been explored for DFDC datasets by any other groups to date. Thus, the proposed model can flexibly accommodate various resolutions and aspect ratios, without being constrained to a specific resolution or aspect ratio for any type of image classification problem. While most of the reported research is limited to sample or preview DFDC datasets only, we have also attempted the testing on full DFDC datasets and presented the results. Contemplating the fact that the detailed results and resource analysis for various scenarios are provided in this research, the proposed deepfake detection method is anticipated to pave new avenues for deepfake detection research, that engages with DFDC datasets.
- Open Access
Semi-Supervised Adversarial Learning Using LSTM for Human Activity Recognition
The training of Human Activity Recognition (HAR) models requires a substantial amount of labeled data. Unfortunately, despite being trained on enormous datasets, most current models have poor performance rates when evaluated against anonymous data from new users. Furthermore, due to the limits and problems of working with human users, capturing adequate data for each new user is not feasible. This paper presents semi-supervised adversarial learning using the LSTM (Long-short term memory) approach for human activity recognition. This proposed method trains annotated and unannotated data (anonymous data) by adapting the semi-supervised learning paradigms on which adversarial learning capitalizes to improve the learning capabilities in dealing with errors that appear in the process. Moreover, it adapts to the change in human activity routine and new activities, i.e., it does not require prior understanding and historical information. Simultaneously, this method is designed as a temporal interactive model instantiation and shows the capacity to estimate heteroscedastic uncertainty owing to inherent data ambiguity. Our methodology also benefits from multiple parallel input sequential data predicting an output exploiting the synchronized LSTM. The proposed method proved to be the best state-of-the-art method with more than 98% accuracy in implementation utilizing the publicly available datasets collected from the smart home environment facilitated with heterogeneous sensors. This technique is a novel approach for high-level human activity recognition and is likely to be a broad application prospect for HAR.
Open AccessSemi-Supervised Adversarial Learning Using LSTM for Human Activity Recognition
The training of Human Activity Recognition (HAR) models requires a substantial amount of labeled data. Unfortunately, despite being trained on enormous datasets, most current models have poor performance rates when evaluated against anonymous data from new users. Furthermore, due to the limits and problems of working with human users, capturing adequate data for each new user is not feasible. This paper presents semi-supervised adversarial learning using the LSTM (Long-short term memory) approach for human activity recognition. This proposed method trains annotated and unannotated data (anonymous data) by adapting the semi-supervised learning paradigms on which adversarial learning capitalizes to improve the learning capabilities in dealing with errors that appear in the process. Moreover, it adapts to the change in human activity routine and new activities, i.e., it does not require prior understanding and historical information. Simultaneously, this method is designed as a temporal interactive model instantiation and shows the capacity to estimate heteroscedastic uncertainty owing to inherent data ambiguity. Our methodology also benefits from multiple parallel input sequential data predicting an output exploiting the synchronized LSTM. The proposed method proved to be the best state-of-the-art method with more than 98% accuracy in implementation utilizing the publicly available datasets collected from the smart home environment facilitated with heterogeneous sensors. This technique is a novel approach for high-level human activity recognition and is likely to be a broad application prospect for HAR.
- Open Access
Epileptic-Net: An Improved Epileptic Seizure Detection System Using Dense Convolutional Block with Attention Network from EEG
Epilepsy is a complex neurological condition that affects a large number of people worldwide. Electroencephalography (EEG) measures the electrical activity of the brain and is widely used in epilepsy diagnosis, but it usually requires manual inspection, which can be hours long, by a neurologist. Several automatic systems have been proposed to detect epilepsy but still have some unsolved issues. In this study, we proposed a dynamic method using a deep learning model (Epileptic-Net) to detect an epileptic seizure. The proposed method is largely heterogeneous and comprised of the dense convolutional blocks (DCB), feature attention modules (FAM), residual blocks (RB), and hypercolumn technique (HT). Firstly, DCB is used to get the discriminative features from the EEG samples. Then, FAM extracts the essential features from the samples. After that, RB learns more vital parts as it entirely uses information in the convolutional layer. Finally, HT retains the efficient local features extracted from the layers situated at the different levels of the model. Its performance has been evaluated on the University of Bonn EEG dataset, divided into five distinct classes. The proposed Epileptic-Net achieves the average accuracy of 99.95% in the two-class classification, 99.98% in the three-class classification, 99.96% in the four-class classification, and 99.96% in classifying the complicated five-class problem. Thus the proposed approach shows more competitive results than the existing model to detect epileptic seizures. We also hope that this method can support experts in achieving objective and reliable results, lowering the misdiagnosis rate, and assisting in decision-making.
Open AccessEpileptic-Net: An Improved Epileptic Seizure Detection System Using Dense Convolutional Block with Attention Network from EEG
Epilepsy is a complex neurological condition that affects a large number of people worldwide. Electroencephalography (EEG) measures the electrical activity of the brain and is widely used in epilepsy diagnosis, but it usually requires manual inspection, which can be hours long, by a neurologist. Several automatic systems have been proposed to detect epilepsy but still have some unsolved issues. In this study, we proposed a dynamic method using a deep learning model (Epileptic-Net) to detect an epileptic seizure. The proposed method is largely heterogeneous and comprised of the dense convolutional blocks (DCB), feature attention modules (FAM), residual blocks (RB), and hypercolumn technique (HT). Firstly, DCB is used to get the discriminative features from the EEG samples. Then, FAM extracts the essential features from the samples. After that, RB learns more vital parts as it entirely uses information in the convolutional layer. Finally, HT retains the efficient local features extracted from the layers situated at the different levels of the model. Its performance has been evaluated on the University of Bonn EEG dataset, divided into five distinct classes. The proposed Epileptic-Net achieves the average accuracy of 99.95% in the two-class classification, 99.98% in the three-class classification, 99.96% in the four-class classification, and 99.96% in classifying the complicated five-class problem. Thus the proposed approach shows more competitive results than the existing model to detect epileptic seizures. We also hope that this method can support experts in achieving objective and reliable results, lowering the misdiagnosis rate, and assisting in decision-making.
- Open Access
Improving R Peak Detection in ECG Signal Using Dynamic Mode Selected Energy and Adaptive Window Sizing Algorithm with Decision Tree Algorithm
R peak detection is crucial in electrocardiogram (ECG) signal analysis to detect and diagnose cardiovascular diseases (CVDs). Herein, the dynamic mode selected energy (DMSE) and adaptive window sizing (AWS) algorithm are proposed for detecting R peaks with better efficiency. The DMSE algorithm adaptively separates the QRS components and all non-objective components from the ECG signal. Based on local peaks in QRS components, the AWS algorithm adaptively determines the Region of Interest (ROI). The Feature Extraction process computes the statistical properties of energy, frequency, and noise from each ROI. The Sequential Forward Selection (SFS) procedure is used to find the best subsets of features. Based on these characteristics, an ensemble of decision tree algorithms detects the R peaks. Finally, the R peak position on the initial ECG signal is adjusted using the R location correction (RLC) algorithm. The proposed method has an experimental accuracy of 99.94%, a sensitivity of 99.98%, positive predictability of 99.96%, and a detection error rate of 0.06%. Given the high efficiency in detection and fast processing speed, the proposed approach is ideal for intelligent medical and wearable devices in the diagnosis of CVDs.
Open AccessImproving R Peak Detection in ECG Signal Using Dynamic Mode Selected Energy and Adaptive Window Sizing Algorithm with Decision Tree Algorithm
R peak detection is crucial in electrocardiogram (ECG) signal analysis to detect and diagnose cardiovascular diseases (CVDs). Herein, the dynamic mode selected energy (DMSE) and adaptive window sizing (AWS) algorithm are proposed for detecting R peaks with better efficiency. The DMSE algorithm adaptively separates the QRS components and all non-objective components from the ECG signal. Based on local peaks in QRS components, the AWS algorithm adaptively determines the Region of Interest (ROI). The Feature Extraction process computes the statistical properties of energy, frequency, and noise from each ROI. The Sequential Forward Selection (SFS) procedure is used to find the best subsets of features. Based on these characteristics, an ensemble of decision tree algorithms detects the R peaks. Finally, the R peak position on the initial ECG signal is adjusted using the R location correction (RLC) algorithm. The proposed method has an experimental accuracy of 99.94%, a sensitivity of 99.98%, positive predictability of 99.96%, and a detection error rate of 0.06%. Given the high efficiency in detection and fast processing speed, the proposed approach is ideal for intelligent medical and wearable devices in the diagnosis of CVDs.
- Open Access
A Deep Machine Learning Method for Concurrent and Interleaved Human Activity Recognition
Human activity recognition has become an important research topic within the field of pervasive computing, ambient assistive living (AAL), robotics, health-care monitoring, and many more. Techniques for recognizing simple and single activities are typical for now, but recognizing complex activities such as concurrent and interleaving activity is still a major challenging issue. In this paper, we propose a two-phase hybrid deep machine learning approach using bi-directional Long-Short Term Memory (BiLSTM) and Skip-Chain Conditional random field (SCCRF) to recognize the complex activity. BiLSTM is a sequential generative deep learning inherited from Recurrent Neural Network (RNN). SCCRFs is a distinctive feature of conditional random field (CRF) that can represent long term dependencies. In the first phase of the proposed approach, we recognized the concurrent activities using the BiLSTM technique, and in the second phase, SCCRF identifies the interleaved activity. Accuracy of the proposed framework against the counterpart state-of-art methods using the publicly available datasets in a smart home environment is analyzed. Our experiment’s result surpasses the previously proposed approaches with an average accuracy of more than 93%.
Open AccessA Deep Machine Learning Method for Concurrent and Interleaved Human Activity Recognition
Human activity recognition has become an important research topic within the field of pervasive computing, ambient assistive living (AAL), robotics, health-care monitoring, and many more. Techniques for recognizing simple and single activities are typical for now, but recognizing complex activities such as concurrent and interleaving activity is still a major challenging issue. In this paper, we propose a two-phase hybrid deep machine learning approach using bi-directional Long-Short Term Memory (BiLSTM) and Skip-Chain Conditional random field (SCCRF) to recognize the complex activity. BiLSTM is a sequential generative deep learning inherited from Recurrent Neural Network (RNN). SCCRFs is a distinctive feature of conditional random field (CRF) that can represent long term dependencies. In the first phase of the proposed approach, we recognized the concurrent activities using the BiLSTM technique, and in the second phase, SCCRF identifies the interleaved activity. Accuracy of the proposed framework against the counterpart state-of-art methods using the publicly available datasets in a smart home environment is analyzed. Our experiment’s result surpasses the previously proposed approaches with an average accuracy of more than 93%.
- 1 March 2020
- Published by Elsevier in Journal of Biomedical Informatics
- Vol. 103, 103371
Unobtrusive monitoring of behavior and movement patterns to detect clinical depression severity level via smartphone
- 1 March 2020
- Published by Elsevier in Journal of Biomedical Informatics
- Vol. 103, 103371
Unobtrusive monitoring of behavior and movement patterns to detect clinical depression severity level via smartphone
- Open Access
- 24 April 2018
- Published by SAGE Publications in International Journal of Distributed Sensor Networks
- Vol. 14 (4)
Log-Viterbi algorithm applied on second-order hidden Markov model for human activity recognition
Recognition of human activities is getting into the limelight among researchers in the field of pervasive computing, ambient intelligence, robotic, and monitoring such as assistive living, elderly care, and health care. Many platforms, models, and algorithms have been developed and implemented to recognize the human activities. However, existing approaches suffer from low-activity accuracy and high time complexity. Therefore, we proposed probabilistic log-Viterbi algorithm on second-order hidden Markov model that facilitates our algorithm by reducing the time complexity with increased accuracy. Second-order hidden Markov model is efficient relevance between previous two activities, current activity, and current observation that incorporate more information into recognition procedure. The log-Viterbi algorithm converts the products of a large number of probabilities into additions and finds the most likely activity from observation sequence under given model. Therefore, this approach maximizes the probability of activity recognition with improved accuracy and reduced time complexity. We compared our proposed algorithm among other famous probabilistic models such as Naïve Bayes, condition random field, hidden Markov model, and hidden semi-Markov model using three datasets in the smart home environment. The recognition possibility of our proposed method is significantly better in accuracy and time complexity than early proposed method. Moreover, this improved algorithm for activity recognition is much effective for almost all the dynamic environments such as assistive living, elderly care, healthcare applications, and home automation.
- 24 April 2018
- Published by SAGE Publications in International Journal of Distributed Sensor Networks
- Vol. 14 (4)
Open AccessLog-Viterbi algorithm applied on second-order hidden Markov model for human activity recognition
Recognition of human activities is getting into the limelight among researchers in the field of pervasive computing, ambient intelligence, robotic, and monitoring such as assistive living, elderly care, and health care. Many platforms, models, and algorithms have been developed and implemented to recognize the human activities. However, existing approaches suffer from low-activity accuracy and high time complexity. Therefore, we proposed probabilistic log-Viterbi algorithm on second-order hidden Markov model that facilitates our algorithm by reducing the time complexity with increased accuracy. Second-order hidden Markov model is efficient relevance between previous two activities, current activity, and current observation that incorporate more information into recognition procedure. The log-Viterbi algorithm converts the products of a large number of probabilities into additions and finds the most likely activity from observation sequence under given model. Therefore, this approach maximizes the probability of activity recognition with improved accuracy and reduced time complexity. We compared our proposed algorithm among other famous probabilistic models such as Naïve Bayes, condition random field, hidden Markov model, and hidden semi-Markov model using three datasets in the smart home environment. The recognition possibility of our proposed method is significantly better in accuracy and time complexity than early proposed method. Moreover, this improved algorithm for activity recognition is much effective for almost all the dynamic environments such as assistive living, elderly care, healthcare applications, and home automation.
- 1 January 2018
- Published by Tech Science Press in Intelligent Automation and Soft Computing
- Vol. 25 (4) , 1-9
State-Space based Linear Modeling for Human Activity Recognition in Smart Space
- 1 January 2018
- Published by Tech Science Press in Intelligent Automation and Soft Computing
- Vol. 25 (4) , 1-9
State-Space based Linear Modeling for Human Activity Recognition in Smart Space