Article
Multidisciplinary Sciences
S. M. Isuru Niroshana, Satoshi Kuroda, Kazuyuki Tanaka, Wenxi Chen
Summary: Timely detection and interpretation of anomalies in an electrocardiogram (ECG) is crucial in healthcare applications. The proposed method utilizes a CNN model with an adaptive windowing algorithm to accurately segment and identify different beats in an ECG signal.
SCIENTIFIC REPORTS
(2023)
Article
Computer Science, Artificial Intelligence
Sean Shensheng Xu, Man-Wai Mak, Chunqi Chang
Summary: This paper proposes an unsupervised patient adaptation approach for creating patient-specific deep neural network (DNN) classifiers based on the patient-specific i-vectors from unlabeled patient-specific ECG data. Evaluation on the MIT-BIH arrhythmia dataset shows that the proposed approach outperforms existing models, making personalized ECG classification more practical.
EXPERT SYSTEMS WITH APPLICATIONS
(2022)
Article
Engineering, Biomedical
Egle Butkeviciute, Liepa Bikulciene, Tomas Blazauskas
Summary: This study investigates the application of continuous non-invasive bio-signal recordings in daily life activities using smart devices and cloud-based technologies. A new ECG feature extraction algorithm for movement-contaminated signals was proposed and validated through comparisons with other methods. This research is significant for real-time data processing and health monitoring.
BIOMEDICAL SIGNAL PROCESSING AND CONTROL
(2022)
Article
Engineering, Biomedical
Liqiang Yuan, Mohammed Yakoob Siyal
Summary: Computer aided diagnosis (CAD) systems based on ECG signals have become essential tools in automated Arrhythmia detection, but they face challenges due to domain shifts and patient privacy concerns. Source free domain adaptation (SFDA) methods using pre-trained models offer a solution to the privacy issue, but class imbalance remains a problem. Therefore, a TAPDA framework is developed to address this issue and achieves better performance than other SFDA methods according to numerical experiments.
BIOMEDICAL SIGNAL PROCESSING AND CONTROL
(2023)
Article
Biotechnology & Applied Microbiology
Xintong Shi, Kohei Yamamoto, Tomoaki Ohtsuki, Yutaka Matsui, Kazunari Owada
Summary: To monitor fetal health and growth, fetal heart rate is a critical indicator. However, the quality of fetal ECG recordings is often affected by noises, making accurate fetal heart rate estimation challenging. This study proposes an unsupervised learning-based approach to assess the quality of fetal ECG signals, achieving a high accuracy in three-level quality classification and reducing errors in fetal heart rate estimation.
BIOENGINEERING-BASEL
(2023)
Article
Computer Science, Information Systems
Abdallah Benhamida, Miklos Kozlovszky
Summary: This paper introduces the basics and applications of electrocardiograms, emphasizing the importance of daily monitoring. It proposes an automated solution for abnormal ECG signal detection and presents an algorithm for ECG pre-annotation and beat-to-beat separation using Autoencoders.
Article
Biology
Rui Hu, Jie Chen, Li Zhou
Summary: This paper proposes a novel transformer-based deep learning neural network, ECG DETR, for arrhythmia detection on continuous single-lead ECG segments. The model simultaneously predicts the positions and categories of all heartbeats within an ECG segment, eliminating the need for explicit heartbeat segmentation. The proposed method shows comparable performance to previous works, achieving high overall accuracy on different arrhythmia detection tasks.
COMPUTERS IN BIOLOGY AND MEDICINE
(2022)
Article
Engineering, Biomedical
Mainul Islam Labib, Abdullah-Al Nahid
Summary: Cardiac arrhythmia refers to irregularities in heartbeats that can cause severe complications if left undiagnosed. Traditional deep learning methods for automated diagnosis often have demographic biases, leading researchers to propose a dynamical systems-based classifier method which achieved high accuracy levels on two benchmark databases.
BIOMEDICAL SIGNAL PROCESSING AND CONTROL
(2022)
Article
Engineering, Biomedical
Marc Junior Nkengue, Xianyi Zeng, Ludovic Koehl, Xuyuan Tao
Summary: Wearable systems and medical image analysis are important tools for COVID-19 monitoring and diagnosis. However, they have limitations. This paper proposes a new wearable system that combines the advantages of these two technologies to achieve real-time monitoring of COVID-19 severity. By introducing a deep neural network model, the proposed system improves diagnostic accuracy and robustness.
BIOMEDICAL SIGNAL PROCESSING AND CONTROL
(2024)
Article
Computer Science, Information Systems
Muhammad Wasimuddin, Khaled Elleithy, Abdelshakour Abuzneid, Miad Faezipour, Omar Abuzaghleh
Summary: Cardiovascular diseases, with Myocardial Infarction being a main focus, are leading causes of mortality globally. Real-time ECG monitoring systems combined with advanced machine learning methods offer health status information, but pose high computing requirements for wearable devices. An improved CNN classifier model is proposed for real-time ECG monitoring of multiple arrhythmia types, achieving an accuracy of 99.23% on validation.
Article
Physiology
Yang Liu, Qince Li, Runnan He, Kuanquan Wang, Jun Liu, Yongfeng Yuan, Yong Xia, Henggui Zhang
Summary: In this work, a weakly supervised deep learning framework for arrhythmia detection (WSDL-AD) is proposed to improve the generalization ability of automatic beat-by-beat arrhythmia detection. By integrating heartbeat classification and recording classification into a deep neural network and utilizing coarsely annotated ECG data, the WSDL-AD framework achieves better performance compared to supervised learning methods. The experimental results demonstrate the potential of this framework for clinical and telehealth applications.
FRONTIERS IN PHYSIOLOGY
(2022)
Article
Computer Science, Information Systems
Ayub Othman Abdulrahman, Karwan Mahdi Hama Rawf, Aree Ali Mohammed
Summary: A model based on the 1D-CNN algorithm is proposed for the binary classification of ECG signals to detect and separate regular and irregular heartbeat signals. Test results on the MIT-BIH dataset showed significant improvements in accuracy, sensitivity, and specificity compared to other related works.
MULTIMEDIA TOOLS AND APPLICATIONS
(2023)
Article
Chemistry, Multidisciplinary
Yuying Liu, Hao Zhang, Kun Zhao, Haiyang Liu, Fei Long, Liping Chen, Yaguang Yang
Summary: This paper presents a new method for assessing the quality of single-channel electrocardiogram (ECG) signals. The method combines the Resnet network structure and the principle of self-attention to extract ECG signal features based on the similarity between individual QRS heartbeats within a ten-second time slice. An improved self-attention module is also introduced in the deep neural network to learn the similarity between features. The results of model testing show that the F1-score can reach 0.954, leading to a more accurate assessment of ECG signal quality.
APPLIED SCIENCES-BASEL
(2023)
Article
Engineering, Electrical & Electronic
L. V. Rajani Kumari, Y. Chalapathi Rao
Summary: In this paper, a Pattern adaptive wavelet-based hybrid approach is proposed for classification of arrhythmia beats. The goal is to categorize Electrocardiogram (ECG) beats into normal and abnormal beats using various machine learning classification methods. Two hybrid classifiers using ensemble learning techniques are proposed to improve the performance. The proposed approach outperforms individual classifiers with increased accuracies.
SIGNAL IMAGE AND VIDEO PROCESSING
(2023)
Article
Biophysics
Emerson Keenan, Chandan Karmakar, Radhagayathri K. Udhayakumar, Fiona C. Brownfoot, Igor Lakhno, Vyacheslav Shulgin, Joachim A. Behar, Marimuthu Palaniswami
Summary: This paper presents a novel method for detecting fetal arrhythmias using short length non-invasive fetal electrocardiography recordings. The method extracts a fetal heart rate time series and computes an entropy profile to classify arrhythmic fetuses. The results demonstrate that this method outperforms other entropy measures and can be used for automated detection of fetal arrhythmias.
PHYSIOLOGICAL MEASUREMENT
(2022)
Article
Engineering, Electrical & Electronic
Shivam Sharma, Udit Satija
Summary: This paper proposes a framework for automated detection and removal of ocular artifacts (OAs) from single-channel EEG signals based on discrete Fourier transform (DFT) and adaptive chirp mode decomposition (ACMD). The proposed framework is evaluated using three publicly available databases and outperforms existing OAs removal techniques.
IEEE SENSORS JOURNAL
(2022)
Article
Engineering, Electrical & Electronic
Neeraj, Udit Satija, Jimson Mathew, R. K. Behera
Summary: In this letter, a unified framework based on attentive cycle-generative adversarial network is proposed for the synthesis of electrocardiogram (ECG) signals from seismocardiogram (SCG) signals. The proposed framework is evaluated on a publicly available database and has shown accurate results in deriving ECG signals from SCG signals. This research has significant practical implications, providing a more comfortable method for patients and assisting in better analysis of cardiac rhythm and arrhythmia.
IEEE SIGNAL PROCESSING LETTERS
(2022)
Article
Computer Science, Information Systems
Gangireddy Narendra Kumar Reddy, M. Sabarimalai Manikandan, N. V. L. Narasimha Murty
Summary: This paper investigates objective distortion measures for quality evaluation of photoplethysmogram (PPG) signals and compares their performance. Evaluation results show that the choice of the best objective distortion measure varies depending on the prediction accuracy and Pearson correlation coefficient in different subjective evaluation groups.
Article
Computer Science, Hardware & Architecture
Ashish Reddy Bommana, Susheel Ujwal Siddamshetty, Dhilleswararao Pudi, K. R. Arvind Thumatti, Srinivas Boppu, M. Sabarimalai Manikandan, Linga Reddy Cenkeramaddi
Summary: This article proposes the design of a vectorized floating-point adder/subtractor that supports arbitrary length floating-point formats. Compared to existing designs in the literature, the proposed design is 2.57x area- and 1.56x power-efficient and supports true vectorization with no restrictions on exponent and mantissa widths.
ACM TRANSACTIONS ON DESIGN AUTOMATION OF ELECTRONIC SYSTEMS
(2023)
Article
Engineering, Electrical & Electronic
Nabasmita Phukan, Shailesh Mohine, Achinta Mondal, M. Sabarimalai Manikandan, Ram Bilas Pachori
Summary: The variation in vital signs may be attributed to daily physical activities rather than organ defects. This study proposes a convolutional neural network-based human activity recognition method and evaluates its performance using acceleration signals from a standard benchmark database. The results demonstrate the importance of selecting optimal hyperparameters and the number of layers to achieve higher accuracy and shorter computational time.
IEEE SENSORS JOURNAL
(2022)
Article
Engineering, Electrical & Electronic
Buduru Naveen Kumar, Srinivas Bhaskar Karanki, M. Sabarimalai Manikandan
Summary: In this paper, a method using dynamic mode decomposition (DMD) is proposed to accurately estimate the amplitudes and frequencies of flicker components. The method includes Hilbert transform for envelope extraction, DMD for spectral analysis, and the assessment of flicker severity index (ΔV10) based on the eye-brain model. The effectiveness of the proposed technique is verified by considering voltage flickers with single and multiple flicker components and investigating sensitive conditions. Simulation results are compared with existing techniques in terms of relative errors of flicker parameters. A hardware prototype is developed for practical flicker measurement and the proposed algorithm is implemented on the Raspberry Pi board, with experimental results compared to simulation findings to prove its applicability.
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
(2023)
Article
Computer Science, Information Systems
Manali Saini, Udit Satija, Madhur Deo Upadhayay
Summary: In this study, a deep learning-based method for mental activity classification using a depthwise separable convolutional neural network with a custom attention unit (DSCNN-CAU) and an IoT implementation using a smartphone for portable brain-computer interface (BCI) applications is proposed. The performance assessment on EEG signals demonstrates high overall accuracies and the evaluation results on smartphone show correct classification of real-time recorded EEG signals. The proposed model and IoT implementation outperform existing techniques in terms of accuracy, robustness against artifacts, latency, and battery current dissipation. Furthermore, cross-database analysis enables the selection of the $F_{P1}$ channel for real-time mental activity classification in IoT-based scenarios.
IEEE INTERNET OF THINGS JOURNAL
(2023)
Article
Engineering, Civil
Aritri Roy, Puneet Kumar Patra, Baidurya Bhattacharya
Summary: Researchers calculated the elastic constants of hydroxyapatite (HAP) using Density Functional Theory (DFT) and molecular dynamics (MD), and found that factors such as temperature and non-bonded interactions affect the elastic response of HAP. The results indicate that the force field based on Buckingham potential performs the best compared to DFT and experimental results.
INTERNATIONAL JOURNAL OF STRUCTURAL STABILITY AND DYNAMICS
(2023)
Article
Engineering, Electrical & Electronic
Sayantika Mandal, Udit Satija
Summary: Due to recent advancements in technology and cost reductions, drones are gaining popularity rapidly. With their increased accessibility, the need for reliable drone detection and identification systems is becoming more critical. In this study, we propose a deep learning model based on a time-frequency multiscale convolutional neural network to detect and identify drones using raw and frequency domain radio frequency signals. Our model outperforms state-of-the-art methods in drone detection and identification using deep neural networks, as evaluated on a publicly accessible database.
IEEE SENSORS LETTERS
(2023)
Article
Engineering, Electrical & Electronic
Arka Roy, Arushi Thakur, Udit Satija
Summary: This letter proposes a method based on visibility graph and residual deep neural network for accurate detection of respiratory sounds, achieving the highest performance rates in a publicly available database.
IEEE SENSORS LETTERS
(2023)
Article
Engineering, Electrical & Electronic
Arka Roy, Udit Satija
Summary: Respiratory diseases are a significant cause of death globally. Early detection is crucial for effective intervention and reducing the spread. This article proposes a lightweight inception network for classifying respiratory diseases using lung sound signals. The proposed framework achieves high accuracy in classifying different pathological classes and outperforms existing works. It provides standardized evaluation and has the potential for real-time clinical applications in automated respiratory health screening.
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
(2023)
Article
Computer Science, Information Systems
Gangireddy Narendra Kumar Reddy, M. Sabarimalai Manikandan, N. V. L. Narasimha Murty, Linga Reddy Cenkeramaddi
Summary: In this paper, a unified quality-aware compression and pulse-respiration rates estimation framework is proposed to reduce energy consumption and false alarms of wearable and edge PPG monitoring devices. By utilizing predictive coding techniques and features extracted from the smoothed prediction error signal, the framework performs signal quality assessment, data compression, and pulse-respiration rates estimation simultaneously.
Article
Engineering, Electrical & Electronic
Arka Roy, Udit Satija
Summary: Chronic obstructive pulmonary disease (COPD) is a major global public health concern. Early detection and accurate diagnosis are crucial for preventing disease progression. Lung sounds provide reliable prognoses for respiratory disease identification. This article proposes a melspectrogram snippet representation learning framework for COPD classification and achieves superior accuracy compared to existing methods.
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
(2023)
Article
Computer Science, Information Systems
Pudi Dhilleswararao, Srinivas Boppu, M. Sabarimalai Manikandan, Linga Reddy Cenkeramaddi
Summary: This paper reviews the research on the development and deployment of DNNs using specialized hardware architectures and embedded AI accelerators. It provides a comparative study of different accelerators based on factors such as power, area, and throughput, and discusses future trends in DNN implementation on specialized hardware accelerators.
Article
Engineering, Electrical & Electronic
Manali Saini, Udit Satija
Summary: In this study, a lightweight 1-D convolutional neural network method is proposed for predicting cognitive activity from electroencephalogram signals. The real-time recorded results demonstrate high prediction accuracy and low power consumption on resource-constrained edge devices.
IEEE SENSORS LETTERS
(2022)
