Article
Engineering, Biomedical
Zhen Liu, Bingyu Zhu, Manfeng Hu, Zhaohong Deng, Jingxiang Zhang
Summary: This paper proposes a revised tunable Q-factor wavelet transform (RTQWT) to overcome the limitations of traditional methods and improve the adaptability to nonstationary EEG signals. Classification experiments using the extracted features show that RTQWT can effectively extract detailed features and improve the classification accuracy of EEG signals.
IEEE TRANSACTIONS ON NEURAL SYSTEMS AND REHABILITATION ENGINEERING
(2023)
Article
Biochemistry & Molecular Biology
Ramy Hussein, Soojin Lee, Rabab Ward
Summary: In this study, a Transformer-based approach called MViT is introduced for automated learning of spatio-temporal-spectral features in multi-channel EEG data. Extensive experiments demonstrate the superiority of MViT algorithm in seizure prediction.
Article
Engineering, Biomedical
Mingkan Shen, Peng Wen, Bo Song, Yan Li
Summary: This paper proposes an EEG based real-time approach using tunable-Q wavelet transform and convolutional neural network (CNN) to detect epilepsy seizures. Statistical moments and spectral band power are used to extract features from EEG and are fed into CNN as imaged-like data. The proposed approach achieves 97.57% accuracy, 98.90% sensitivity, 2.13% false positive rate and a delay of 10.46 seconds. It is also suitable for real-time implementation and can be applied to clinical seizure detection.
BIOMEDICAL SIGNAL PROCESSING AND CONTROL
(2023)
Article
Computer Science, Information Systems
Qi Xin, Shaohao Hu, Shuaiqi Liu, Ling Zhao, Shuihua Wang
Summary: A method named WTRPNet, a explainable graph feature convolutional neural network, is proposed for epileptic EEG classification, showing superior performance. Experimental results demonstrate an accuracy of 99.67% in the classification of focal and nonfocal epileptic EEG, proving its effectiveness in classification and detection.
ACM TRANSACTIONS ON MULTIMEDIA COMPUTING COMMUNICATIONS AND APPLICATIONS
(2021)
Article
Computer Science, Artificial Intelligence
Ramy Hussein, Soojin Lee, Rabab Ward, Martin J. McKeown
Summary: This study introduces a novel semi-dilated convolutional neural network architecture that outperforms previous methods in predicting epileptic seizures, achieving an average prediction sensitivity of 98.90% for scalp EEG.
Article
Mathematics, Interdisciplinary Applications
Arshpreet Kaur, Kumar Shashvat
Summary: This study aims to automate the identification of inter-ictal activity from EEG and distinguish it from the activity of a controlled patient. The researchers used the Bonn dataset and patient data collected from Max Hospital, Saket, and applied Continuous Wavelet Transform and a fifteen-layer Convolutional Neural Network for signal classification. The results show that the proposed method outperforms existing methods in terms of performance metrics, and scalograms are effective for identifying epileptic states.
CHAOS SOLITONS & FRACTALS
(2022)
Article
Engineering, Biomedical
Mingyang Li, Wanzhong Chen, Min Xia
Summary: This paper presents a robust feature extraction scheme for analyzing epileptic EEG using synchrosqueezing short-time Fourier transform (SSTFT) and graph regularized non-negative matrix factorization (GNMF). The proposed method achieves high accuracy in discriminating seizure events and offers a new perspective for the design of automatic EEG monitoring systems.
BIOMEDICAL SIGNAL PROCESSING AND CONTROL
(2023)
Article
Computer Science, Artificial Intelligence
Ibrahim Aliyu, Chang Gyoon Lim
Summary: This paper proposes an LSTM network for classifying epileptic EEG signals. Discrete wavelet transform is used to remove noise and extract features, and the optimal features are identified through correlation and P value analysis. The proposed method achieves high accuracy and outperforms other popular classifiers.
NEURAL COMPUTING & APPLICATIONS
(2023)
Article
Engineering, Biomedical
Abdulhamit Subasi, Turker Tuncer, Sengul Dogan, Dahiru Tanko, Unal Sakoglu
Summary: The paper introduces a new automated emotion recognition framework using EEG signals, achieving over 93% classification accuracy through various processing methods, demonstrating its effectiveness in emotion recognition.
BIOMEDICAL SIGNAL PROCESSING AND CONTROL
(2021)
Article
Computer Science, Information Systems
Varsha Harpale, Vinayak Bairagi
Summary: EEG analysis plays a crucial role in detecting and predicting various brain diseases, with a focus on classifying normal EEG signals from epileptic EEG signals. The study aims to identify pre-seizure and seizure states of EEG signals using time and frequency features, utilizing a fuzzy classifier for prediction accuracy.
JOURNAL OF KING SAUD UNIVERSITY-COMPUTER AND INFORMATION SCIENCES
(2021)
Article
Engineering, Biomedical
Chandan Kumar Jha, Maheshkumar H. Kolekar
Summary: The paper presents a novel ECG data compression technique based on the tunable Q-wavelet transform, achieving good compression performance by compacting signal energy and discarding small valued transform coefficients. Experimental results show that the proposed technique performs well in cardiac arrhythmia classification.
BIOMEDICAL SIGNAL PROCESSING AND CONTROL
(2021)
Article
Computer Science, Artificial Intelligence
Xiaoying Wang, Yu Ling, Xiang Ling, Xianghuan Li, Zhicheng Li, Kunpeng Hu, Min Dai, Jia Zhu, Yuxiao Du, Qintai Yang
Summary: This study decomposes the EEG signal of patients into multiple intrinsic modal functions using empirical modal decomposition, and classifies them using a fusion algorithm of support vector machine and K-nearest neighbor optimized by particle swarm algorithm. The validation results show that the proposed method can achieve a high accuracy recognition rate of epileptic seizures and has good clinical application value.
INTERNATIONAL JOURNAL OF INTELLIGENT SYSTEMS
(2022)
Article
Engineering, Biomedical
Tao Zhang, Zhiwu Han, Xiaojuan Chen, Wanzhong Chen
Summary: A novel fusion method combining FSWT, FuzzyEn, HFD, t-SNE, and KNN was proposed for automated detection of epileptic EEG. Experimental results showed high accuracies of the proposed method in seizure detection, indicating that nonlinear features based on subbands and CSoS enhance performance.
BIOMEDICAL SIGNAL PROCESSING AND CONTROL
(2021)
Article
Engineering, Multidisciplinary
Sepehr Nouhi, Masoud Pour
Summary: The paper presents a technique combining surface photography and wavelet method to predict future surface roughness by extracting time delay parameters, embedding dimension, and false nearest neighbor parameters. The study shows that this method can be applied in various machining processes with a consistent prediction error of about 7% after Ra = 0.4 μm in different processes.
Review
Chemistry, Multidisciplinary
Federico Fogolari, Roberto Borelli, Agostino Dovier, Gennaro Esposito
Summary: The article introduces the application and advantages of the kth nearest neighbor method in entropy estimation, as well as the relevant variables, metrics, and applications associated with this method. By combining this method with mutual information, high-dimensional problems can be addressed.
WILEY INTERDISCIPLINARY REVIEWS-COMPUTATIONAL MOLECULAR SCIENCE
(2023)
Article
Computer Science, Artificial Intelligence
U. Raghavendra, Anjan Gudigar, Yashas Chakole, Praneet Kasula, D. P. Subha, Nahrizul Adib Kadri, Edward J. Ciaccio, U. Rajendra Acharya
Summary: This study proposes a method for diagnosing depression using machine learning and continuous wavelet transform. By extracting and reducing features from electroencephalogram recordings, and labeling them using various classifiers, high diagnostic accuracies were achieved. Additionally, a depression severity index was developed for distinguishing between normal and depressed classes.
Article
Computer Science, Artificial Intelligence
Manish Sharma, Sohamkumar Patel, U. Rajendra Acharya
Summary: Congestive heart failure (CHF) is a cardiac disorder caused by inefficient pumping of the heart, resulting in insufficient blood flow. This study proposes a method using an optimized wavelet filter bank and heart rate variability (HRV) signals to automatically identify CHF. The method achieves high classification accuracy when using classifiers such as support vector machine (SVM).
Article
Neurosciences
Afshin Shoeibi, Navid Ghassemi, Marjane Khodatars, Parisa Moridian, Abbas Khosravi, Assef Zare, Juan M. Gorriz, Amir Hossein Chale-Chale, Ali Khadem, U. Rajendra Acharya
Summary: This paper presents an intelligent detection method for SZ and ADHD using a new deep learning approach based on rs-fMRI data. The proposed method employs an IT2FR fuzzy method for classification, optimized by the GWO algorithm, achieving satisfactory results.
COGNITIVE NEURODYNAMICS
(2023)
Article
Computer Science, Interdisciplinary Applications
V. Jahmunah, E. Y. K. Ng, Ru-San Tan, Shu Lih Oh, U. Rajendra Acharya
Summary: A Dirichlet DenseNet model was developed to analyze out-of-distribution data and detect misclassification of myocardial infarction (MI) and normal electrocardiogram (ECG) signals. The model was trained with pre-processed MI ECG signals from the PTB database and showed increased confidence in classifying signals with lower levels of noise. The model proved reliable in diagnosing MI.
COMPUTER METHODS AND PROGRAMS IN BIOMEDICINE
(2023)
Article
Biology
V. Jahmunah, Sylvia Chen, Shu Lih Oh, U. Rajendra Acharya, Balram Chowbay
Summary: Existing warfarin dose prediction algorithms based on pharmacogenetics and clinical parameters have not been used clinically due to lack of external validation, assessment for clinical utility, and high risk of bias. Deep neural models can improve the precision and accuracy of warfarin dose predictions, but heterogeneity across datasets and ethnic populations affects dosing accuracy. This study developed a deep learning model using a well-established dataset to predict warfarin dose, resulting in low mean absolute error and percentage of error in different ethnic populations.
COMPUTERS IN BIOLOGY AND MEDICINE
(2023)
Review
Biophysics
Manish Sharma, Ruchit Kumar Patel, Akshat Garg, Ru SanTan, U. Rajendra Acharya
Summary: Schizophrenia is a devastating mental disorder that affects higher brain functions and has a profound impact on individuals. Deep learning models can automatically detect schizophrenia by learning signal data characteristics, without the need for traditional feature engineering. This systematic review explores various deep learning models and methodologies for schizophrenia detection based on EEG signals, structural and functional MRI data from diverse datasets. The study discusses the challenges and future works in using deep learning models for schizophrenia diagnosis.
PHYSIOLOGICAL MEASUREMENT
(2023)
Review
Chemistry, Analytical
Roohallah Alizadehsani, Mohamad Roshanzamir, Navid Hoseini Izadi, Raffaele Gravina, H. M. Dipu Kabir, Darius Nahavandi, Hamid Alinejad-Rokny, Abbas Khosravi, U. Rajendra Acharya, Saeid Nahavandi, Giancarlo Fortino
Summary: Continuous advancements in technologies like the internet of things and big data analysis have enabled information sharing and smart decision-making using everyday devices. Swarm intelligence algorithms facilitate constructive interaction among individuals regardless of their intelligence level to address complex nonlinear problems. This paper examines the application of swarm intelligence algorithms in the internet of medical things, with a focus on wearable devices in healthcare. It reviews existing works on utilizing swarm intelligence in tackling IoMT problems such as disease prediction, data encryption, and resource allocation. The paper concludes with research perspectives and future trends.
Review
Computer Science, Artificial Intelligence
Joshua Sheehy, Hamish Rutledge, U. Rajendra Acharya, Hui Wen Loh, Raj Gururajan, Xiaohui Tao, Xujuan Zhou, Yuefeng Li, Tiana Gurney, Srinivas Kondalsamy-Chennakesavan
Summary: This systematic review aims to evaluate and critique the methodologies and approaches used in predicting the prognosis of gynecological cancers using machine learning techniques. A total of 139 studies met the inclusion criteria, with varying study quality and inconsistent methodologies, statistical reporting, and outcome measures. This hinders the ability to perform meta-analysis and draw conclusions regarding the superiority of machine learning methods.
ARTIFICIAL INTELLIGENCE IN MEDICINE
(2023)
Review
Computer Science, Artificial Intelligence
Afshin Shoeibi, Marjane Khodatars, Mahboobeh Jafari, Navid Ghassemi, Parisa Moridian, Roohallah Alizadehsani, Sai Ho Ling, Abbas Khosravi, Hamid Alinejad-Rokny, H. K. Lam, Matthew Fuller-Tyszkiewicz, U. Rajendra Acharya, Donovan Anderson, Yudong Zhang, Juan Manuel Gorriz
Summary: Brain diseases, including tumors and mental and neurological disorders, pose serious threats to the health and well-being of millions of people worldwide. Neuroimaging modalities are commonly used by physicians to aid the diagnosis of brain diseases. However, fusing these modalities for accurate diagnosis can be challenging for specialist doctors.
INFORMATION FUSION
(2023)
Article
Automation & Control Systems
Manish Sharma, Divyansh Anand, Sarv Verma, U. Rajendra Acharya
Summary: Sleep is crucial for human well-being, and insomnia is a common sleep disorder that affects both physical and mental health. This study proposes a method that uses single-channel EEG signals to automatically identify insomnia, extracting features using a deep convolutional network and developing a model for sleep stages classification.
ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE
(2023)
Article
Computer Science, Artificial Intelligence
Erhan Akbal, Prabal Datta Barua, Sengul Dogan, Turker Tuncer, U. Rajendra Acharya
Summary: Classification of animal species using animal sounds is important for bioacoustics work, especially for determining anurans as an indicator of climate change. However, it is challenging to count and classify anurans in their natural habitat, thus requiring computer-assisted intelligent systems. This study collected a new anuran sound dataset and proposed a hand-modeled sound classification system, achieving a high classification accuracy of 99.35% for 26 anuran species using an improved 1D-LBP and TQWT-based feature extraction method.
EXPERT SYSTEMS WITH APPLICATIONS
(2023)
Review
Computer Science, Artificial Intelligence
Smith K. Khare, Sonja March, Prabal Datta Barua, Vikram M. Gadre, U. Rajendra Acharya
Summary: Mental health is essential for a sustainable and developing society. The prevalence and financial burden of mental illness have increased globally, and this paper provides a systematic review of nine developmental and mental disorders in children and adolescents. The paper focuses on the use of physiological signals for automated detection of these disorders, and discusses signal analysis, feature engineering, decision-making, challenges, and future directions in this field. The main findings of the study are presented in the conclusion section.
INFORMATION FUSION
(2023)
Article
Computer Science, Artificial Intelligence
Moloud Abdar, Arash Mehrzadi, Milad Goudarzi, Farzad Masoudkabir, Leonardo Rundo, Mohammad Mamouei, Evis Sala, Abbas Khosravi, Vladimir Makarenkov, U. Rajendra Acharya, Seyedmohammad Saadatagah, Mohammadreza Naderian, Salvador Garcia, Nizal Sarrafzadegan, Saeid Nahavandi
Summary: In this study, machine learning methods combined with uncertainty quantification were used to accurately classify CSX data from the coronary angiography registry of Tehran's Heart Center. The proposed model reached an accuracy of 85% when applied to the benchmark CSX dataset.
INFORMATION FUSION
(2023)
Article
Computer Science, Information Systems
Prabal Datta Barua, Makiko Kobayashi, Masayuki Tanabe, Mehmet Baygin, Jose Kunnel Paul, Thomas Iype, Sengul Dogan, Turker Tuncer, Ru-San Tan, U. Rajendra Acharya
Summary: This study aimed to develop a lightweight machine learning model for diagnosing fibromyalgia using single-lead ECG signals recorded during sleep. The model achieved high accuracy in differentiating the electrocardiographic signatures of fibromyalgia patients from control subjects, demonstrating its efficacy in clinical integration.
Article
Computer Science, Artificial Intelligence
Sinan Tatli, Gulay Macin, Irem Tasci, Burak Tasci, Prabal Datta Barua, Mehmet Baygin, Turker Tuncer, Sengul Dogan, Edward J. Ciaccio, U. Rajendra Acharya
Summary: This study aims to propose a new algorithm for early diagnosis of multiple sclerosis (MS) using machine learning. The algorithm utilizes transfer learning and hybrid feature engineering, and calculates feature vectors through multiple layers of neural networks, resulting in high classification accuracy.
EXPERT SYSTEMS WITH APPLICATIONS
(2024)