Article
Engineering, Biomedical
Jefferson Tales Oliva, Joao Luis Garcia Rosa
Summary: A novel method for epilepsy detection using binary and multiclass classifiers was presented and validated on an EEG database. BP-MLP and SMO_Pol algorithms showed the highest accuracy for binary and multiclass classification problems.
BIOMEDICAL SIGNAL PROCESSING AND CONTROL
(2021)
Article
Engineering, Biomedical
Siuly Siuly, Yanhui Guo, Omer Faruk Alcin, Yan Li, Peng Wen, Hua Wang
Summary: This study introduces a feature extraction method based on deep residual networks to automatically extract features from EEG signal data for diagnosing schizophrenia. The experimental results demonstrate that this method outperforms existing approaches and can discover important biomarkers, aiding in the development of a computer-assisted diagnostic system.
PHYSICAL AND ENGINEERING SCIENCES IN MEDICINE
(2023)
Review
Chemistry, Analytical
Kais Belwafi, Sofien Gannouni, Hatim Aboalsamh
Summary: BCI systems have a wide range of applications in restoring capabilities for people with severe motor disabilities, with a growing number of systems being developed. There is a significant interest in implementing BCIs on portable platforms, with smaller size, faster loading times, lower cost, fewer resources, and lower power consumption compared to full PCs.
Article
Medicine, General & Internal
Prabal Datta Barua, Sengul Dogan, Mehmet Baygin, Turker Tuncer, Elizabeth Emma Palmer, Edward J. Ciaccio, U. Rajendra Acharya
Summary: In this research, a hand-modeled classification model using a new ternary motif pattern (TMP) has been proposed for separating healthy versus ADHD individuals based on noisy EEG signals. The model utilizes the Tunable Q Wavelet Transform (TQWT) for feature extraction and applies neighborhood component analysis (NCA) and k-nearest neighbor (kNN) classifier for feature selection and classification. The model achieved high classification accuracies of 95.57% and 77.93% using 10-fold and leave one subject out (LOSO) cross-validations, respectively.
Article
Chemistry, Multidisciplinary
Uriel Calderon-Uribe, Rocio A. Lizarraga-Morales, Igor V. Guryev
Summary: The detection of faults in induction motors is a major challenge for the industry. This paper proposes a method for the detection of an unbalance fault in induction motors using a low-dimensional feature vector and a low-complexity classification approach. The method utilizes texture features to analyze vibration signals from unbalanced and healthy motors. Experimental results show higher accuracy and lower training time compared to state-of-the-art approaches.
APPLIED SCIENCES-BASEL
(2023)
Article
Computer Science, Information Systems
Aayesha, Muhammad Bilal Qureshi, Muhammad Afzaal, Muhammad Shuaib Qureshi, Muhammad Fayaz
Summary: This paper focuses on extracting distinguishing features of seizure EEG recordings to develop an approach that employs both fuzzy-based and traditional machine learning algorithms for epileptic seizure detection. The obtained results show that K-Nearest Neighbor (KNN) and Fuzzy Rough Nearest Neighbor (FRNN) give the highest classification accuracy scores, with improved sensitivity and specificity percentages.
MULTIMEDIA TOOLS AND APPLICATIONS
(2021)
Article
Computer Science, Artificial Intelligence
Hesam Akbari, Muhammad Tariq Sadiq, Malih Payan, Somayeh Saraf Esmaili, Hourieh Baghri, Hamed Bagheri
Summary: This research introduces a novel strategy for diagnosing depression based on geometric features derived from EEG signal shape, utilizing Binary Particle Swarm Optimization for feature selection and support vector machine and K-nearest neighbor classifiers for signal identification. The proposed system achieves an average classification accuracy of 98.79% in a study involving 22 normal and 22 depressed subjects.
TRAITEMENT DU SIGNAL
(2021)
Article
Computer Science, Artificial Intelligence
Wei Yan Peh, John Thomas, Elham Bagheri, Rima Chaudhari, Sagar Karia, Rahul Rathakrishnan, Vinay Saini, Nilesh Shah, Rohit Srivastava, Yee-Leng Tan, Justin Dauwels
Summary: The study proposed three automated approaches to detect slowing in EEG, with Deep Learning-based Detection System achieving the best overall results in terms of accuracy.
INTERNATIONAL JOURNAL OF NEURAL SYSTEMS
(2021)
Article
Multidisciplinary Sciences
Roberto Barcenas, Ruth Fuentes-Garcia, Lizbeth Naranjo
Summary: This article explains the Unified Parkinson's Disease Rating Scale (UPDRS) as a measure of disease progression in Parkinson's disease using voice signals. The introduction of a Support Vector Regression (SVR) model based on a combination of kernel functions yields significant improvements compared to other learning approaches.
Review
Computer Science, Artificial Intelligence
Niamh McCallan, Scot Davidson, Kok Yew Ng, Pardis Biglarbeigi, Dewar Finlay, Boon Leong Lan, James McLaughlin
Summary: This paper introduces how epilepsy affects people, and presents a method for utilizing frequency and amplitude information from multiple seizure types to aid in the development of future seizure classification algorithms. Through a detailed review and analysis of the Temple University Hospital Seizure Corpus, it is found that deep learning techniques perform the best in seizure classification. Finally, the limitations of the TUSZ dataset are highlighted and future work suggestions are provided.
EXPERT SYSTEMS WITH APPLICATIONS
(2023)
Article
Computer Science, Artificial Intelligence
Hong Peng, Cancheng Li, Jinlong Chao, Tao Wang, Chengjian Zhao, Xiaoning Huo, Bin Hu
Summary: This study proposes a novel sparse representation-based epileptic seizure classification method based on dictionary learning, which is evaluated on public EEG databases. The new method shows higher automation and recognition rates compared to traditional methods.
Article
Engineering, Biomedical
Cameron J. Huggins, Javier Escudero, Mario A. Parra, Brian Scally, Renato Anghinah, Amanda Vitoria Lacerda De Araujo, Luis F. Basile, Daniel Abasolo
Summary: This study presents a novel DL model for classifying AD, MCI, and HA subjects using resting-state scalp EEG signals. By preprocessing, continuous wavelet transform, and DL model, a high classification accuracy was achieved.
JOURNAL OF NEURAL ENGINEERING
(2021)
Article
Physics, Multidisciplinary
Egils Avots, Klavs Jermakovs, Maie Bachmann, Laura Paeske, Cagri Ozcinar, Gholamreza Anbarjafari
Summary: This research aims to determine the long-lasting effects of depression through EEG signals. After comparing several classifiers and feature selection methods, the results show that the EEG features used for classifying ongoing depression also work for classifying the long-lasting effects of depression.
Article
Biochemical Research Methods
Hui Bi, Shumei Cao, Hanying Yan, Zhongyi Jiang, Jun Zhang, Ling Zou
Summary: Monitoring the depth of anesthesia is important for administering general anesthetics during surgery. However, traditional EEG monitors have limitations in monitoring conscious states. This study used high-density EEG signals to compare two methods for functional connectivity analysis before and after anesthesia-induced loss of consciousness. The results show that the method based on sparse representation performs better in distinguishing loss of consciousness from awake states.
IEEE-ACM TRANSACTIONS ON COMPUTATIONAL BIOLOGY AND BIOINFORMATICS
(2022)
Article
Computer Science, Artificial Intelligence
Souhir Khessiba, Ahmed Ghazi Blaiech, Khaled Ben Khalifa, Asma Ben Abdallah, Mohamed Hedi Bedoui
Summary: Electroencephalography (EEG) is commonly used for studying brain electrical activity. Deep learning networks can accurately predict individuals' states of vigilance based on EEG signals. Experimental results demonstrate that the proposed 1D-UNet and 1D-UNet-LSTM models perform well in stabilizing training and recognizing vigilance states, indicating the effectiveness of the proposed methods.
NEURAL COMPUTING & APPLICATIONS
(2021)
Article
Chemistry, Analytical
Md Shafayet Hossain, Muhammad E. H. Chowdhury, Mamun Bin Ibne Reaz, Sawal Hamid Md Ali, Ahmad Ashrif A. Bakar, Serkan Kiranyaz, Amith Khandakar, Mohammed Alhatou, Rumana Habib, Muhammad Maqsud Hossain
Article
Chemistry, Analytical
Amith Khandakar, Muhammad E. H. Chowdhury, Mamun Bin Ibne Reaz, Sawal Hamid Md Ali, Serkan Kiranyaz, Tawsifur Rahman, Moajjem Hossain Chowdhury, Mohamed Arselene Ayari, Rashad Alfkey, Ahmad Ashrif A. Bakar, Rayaz A. Malik, Anwarul Hasan
Summary: Diabetic foot ulcers are a serious complication for diabetic patients, and infrared foot thermograms can be used to detect ulceration risk. This study utilizes machine learning techniques and expert validation to classify thermograms, with the VGG 19 CNN model achieving the best performance in severity stratification.
Article
Mathematical & Computational Biology
Md. Johirul Islam, Shamim Ahmad, Fahmida Haque, Mamun Bin Ibne Reaz, Mohammad A. S. Bhuiyan, Khairun Nisa' Minhad, Md. Rezaul Islam
Summary: This study proposes two time-domain features based on nonlinear scaling, LMAV and NSV, for electromyogram (EMG) pattern recognition. Experimental results show that the proposed feature extraction method improves the performance of EMG pattern recognition.
COMPUTATIONAL INTELLIGENCE AND NEUROSCIENCE
(2022)
Article
Mathematical & Computational Biology
Fahmida Haque, Mamun B. I. Reaz, Muhammad E. H. Chowdhury, Serkan Kiranyaz, Sawal H. M. Ali, Mohammed Alhatou, Rumana Habib, Ahmad A. A. Bakar, Norhana Arsad, Geetika Srivastava
Summary: This study explored the use of machine learning models in the diagnosis of diabetic sensorimotor polyneuropathy (DSPN). The results showed that the ensemble classifier and random forest model performed well in diagnosing DSPN using nerve conduction studies data, and can enhance the management of DSPN patients.
COMPUTATIONAL INTELLIGENCE AND NEUROSCIENCE
(2022)
Article
Chemistry, Analytical
Amith Khandakar, Sakib Mahmud, Muhammad E. H. Chowdhury, Mamun Bin Ibne Reaz, Serkan Kiranyaz, Zaid Bin Mahbub, Sawal Hamid Md Ali, Ahmad Ashrif A. Bakar, Mohamed Arselene Ayari, Mohammed Alhatou, Mohammed Abdul-Moniem, Md Ahasan Atick Faisal
Summary: An intelligent insole system is designed to monitor individuals' foot pressure and temperature in real-time. It utilizes off-the-shelf sensors to detect plantar pressure and temperature, and data can be wirelessly transmitted to a centralized device for storage. The research aims to create an affordable, practical, and portable foot monitoring system for continuous at-home monitoring of foot problems and early detection of diabetic foot complications.
Article
Engineering, Electrical & Electronic
Amith Khandakar, Muhammad E. H. Chowdhury, Mamun Bin Ibne Reaz, Serkan Kiranyaz, Anwarul Hasan, Tawsifur Rahman, Sawal Hamid Md. Ali, Mohd Ibrahim bin Shapiai Abd Razak, Ahmad Ashrif A. Bakar, Kanchon Kanti Podder, Moajjem Hossain Chowdhury, Md. Ahasan Atick Faisal, Rayaz A. Malik
Summary: Diabetic sensorimotor polyneuropathy (DSPN) can result in pain, diabetic foot ulceration (DFU), amputation, and death. This study proposes a robust machine-learning approach called DSPNet to identify patients with severe DSPN using standing foot temperature maps. The study achieved an F1 score of 90.3% and outperformed current deep-learning network methods, indicating the effectiveness of temperature maps in detecting high-risk DFU patients and identifying severe DSPN patients. Such sensors can be easily incorporated into smart insoles.
IEEE SENSORS JOURNAL
(2023)
Article
Medicine, General & Internal
Fahmida Haque, Mamun B. I. Reaz, Muhammad E. H. Chowdhury, Mohd Ibrahim bin Shapiai, Rayaz S. A. Malik, Mohammed Alhatou, Syoji Kobashi, Iffat Ara, Sawal H. M. Ali, Ahmad A. A. Bakar, Mohammad Arif Sobhan Bhuiyan
Summary: Diabetic sensorimotor polyneuropathy (DSPN) is a serious complication of diabetes that can lead to foot ulceration and amputation. The Michigan neuropathy screening instrument (MNSI) is commonly used for screening DSPN but lacks a measure of severity. In this study, a DSPN severity grading system based on machine learning algorithms was developed using longitudinal data from the EDIC trial. The system utilizes the top seven MNSI features, including vibration perception, filament test, previous neuropathy, callus, deformities, and fissure, to detect DSPN severity. The developed nomogram showed high accuracy in predicting DSPN and can be used to determine prognosis in patients with DSPN.
Article
Biotechnology & Applied Microbiology
Moajjem Hossain Chowdhury, Md Nazmul Islam Shuzan, Muhammad E. H. Chowdhury, Mamun Bin Ibne Reaz, Sakib Mahmud, Nasser Al Emadi, Mohamed Arselene Ayari, Sawal Hamid Md Ali, Ahmad Ashrif A. Bakar, Syed Mahfuzur Rahman, Amith Khandakar
Summary: This paper proposes a deep-learning-based solution to estimate RR directly from the PPG signal, providing a new method for monitoring respiratory ailments. The lightweight model outperforms other deep neural networks and shows promising results on different datasets. It can be deployed to mobile devices for real-time monitoring.
BIOENGINEERING-BASEL
(2022)
Article
Biotechnology & Applied Microbiology
Md Shafayet Hossain, Sakib Mahmud, Amith Khandakar, Nasser Al-Emadi, Farhana Ahmed Chowdhury, Zaid Bin Mahbub, Mamun Bin Ibne Reaz, Muhammad E. H. Chowdhury
Summary: This paper proposes a novel one-dimensional convolutional neural network (1D-CNN) called MultiResUNet3+ to remove physiological artifacts from electroencephalogram (EEG) signals. A publicly available dataset is used to train, validate, and test the proposed model along with four other 1D-CNN models. The results show that MultiResUNet3+ achieves the highest reduction in EOG and EMG artifacts compared to the other models.
BIOENGINEERING-BASEL
(2023)
Review
Medicine, General & Internal
Khandaker Reajul Islam, Johayra Prithula, Jaya Kumar, Toh Leong Tan, Mamun Bin Ibne Reaz, Md. Shaheenur Islam Sumon, Muhammad E. H. Chowdhury
Summary: This systematic review examines the application of machine learning and deep learning in predicting sepsis using electronic health records. The study highlights the importance of these methods in early sepsis detection and improving patient outcomes.
JOURNAL OF CLINICAL MEDICINE
(2023)
Article
Engineering, Electrical & Electronic
Charn Loong Ng, Mamun Bin Ibne Reaz, Maria Liz Crespo, Andres Cicuttin, Mohd Ibrahim Bin Shapiai, Sawal Hamid Bin Md Ali, Noorfazila Binti Kamal, Muhammad Enamul Hoque Chowdhury
Summary: Musculoskeletal diseases have a negative impact on personal health and the global economy. Wearable sensing technology, specifically the capacitive EMG biomedical sensor introduced in this research, can improve the efficiency of public healthcare strategies to combat these diseases. The sensor's flexibility, miniaturization, and durability make it suitable for standalone adhesive use or integration into wearable applications.
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
(2023)
Article
Engineering, Electrical & Electronic
Md. Johirul Islam, Shamim Ahmad, Fahmida Haque, Mamun Bin Ibne Reaz, Mohammad Arif Sobhan Bhuiyan, Md. Rezaul Islam
Summary: Surface electromyography (EMG) is a promising signal for hand movement recognition. However, subject-dependent EMG pattern recognition limits its use for different subjects. This study proposes a subject invariant EMG pattern recognition method by extracting subject invariant features and using spectral regression discriminant analysis (SRDA) for dimensionality reduction. The proposed method achieves high F1 scores and outperforms subject independent and subject-dependent methods, while being simple, classifier independent, and time complexity free.
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
(2022)
Article
Chemistry, Analytical
Fahmida Haque, Mamun Bin Ibne Reaz, Muhammad Enamul Hoque Chowdhury, Maymouna Ezeddin, Serkan Kiranyaz, Mohammed Alhatou, Sawal Hamid Md Ali, Ahmad Ashrif A. Bakar, Geetika Srivastava
Summary: This study proposes the use of machine learning techniques to identify DN and DFU patients using EMG and GRF data. The KNN algorithm performed well in identifying DN and DFU, achieving high accuracy when optimized.
Article
Computer Science, Information Systems
Md Johirul Islam, Shamim Ahmad, Fahmida Haque, Mamun Bin Ibne Reaz, Mohammad A. S. Bhuiyan, Md Rezaul Islam
Summary: This study proposes a feature selection method for improving electromyogram pattern recognition in prosthetic hands. The selected features achieve significant improvements in accuracy and F1 score, and the method achieves forearm orientation and muscle force invariant performance in training the classifier.
Article
Computer Science, Information Systems
Md Shafayet Hossain, Mamun Bin Ibne Reaz, Muhammad E. H. Chowdhury, Sawal H. M. Ali, Ahmad Ashrif A. Bakar, Serkan Kiranyaz, Amith Khandakar, Mohammed Alhatou, Rumana Habib
Summary: Physiological signal measurement and processing are gaining popularity in ambulatory settings. This paper proposes three novel multiresolution analysis techniques for motion artifact correction from EEG and fNIRS signals. The results show that these methods outperform existing techniques in denoising performance.