Prediction of biomedical signals using deep learning techniques

Article type: Research Article

Authors: Kalaivani, K.^{a; *} | Kshirsagarr, Pravin R.^b | Sirisha Devi, J.^c | Bandela, Surekha Reddy^d | Colak, Ilhami^e | Nageswara Rao, J.^f | Rajaram, A.^g

Affiliations: [a] Department of School of Computer Science and Engineering, Vellore Institute of Technology, Vellore | [b] Department of Artificial Intelligence, G. H. Raisoni College of Engineering, Nagpur, India | [c] Department of Computer Science and Engineering, Institute of Aeronautical Engineering Dundigal, Hyderabad, Telangana, India | [d] Department of ECE, Institute of Aeronautical Engineering, Hyderabad, India | [e] Nisantasi University, Engineering and Architecture Faculty, Department of Electrical and Electronics Engineering, Istanbul, Turkiye | [f] Lakireddy Bali Reddy College of Engineering, Andhra Pradesh, India | [g] Department of Electronics and Communication Engineering E.G.S Pillay Engineering College, Nagapattinam, Tamil Nadu, India

Correspondence: [*] Corresponding author. K. Kalaivani, Assistant Professor Senior Grade1, Department of School of Computer Science and Engineering, Vellore Institute of Technology, Vellore, 632014, E-mail: kalaivani8988@gmail.com.

Abstract: The electrocardiogram (ECG), electroencephalogram (EEG), and electromyogram (EMG) are all very useful diagnostic techniques. The widespread availability of mobile devices plus the declining cost of ECG, EEG, and EMG sensors provide a unique opportunity for making this kind of study widely available. The fundamental need for enhancing a country’s healthcare industry is the ability to foresee the plethora of ailments with which people are now being diagnosed. It’s no exaggeration to say that heart disease is one of the leading causes of mortality and disability in the world today. Diagnosing heart disease is a difficult process that calls for much training and expertise. Electrocardiogram (ECG) signal is an electrical signal produced by the human heart and used to detect the human heartbeat. Emotions are not simple phenomena, yet they do have a major impact on the standard of living. All of these mental processes including drive, perception, cognition, creativity, focus, attention, learning, and decision making are greatly influenced by emotional states. Electroencephalogram (EEG) signals react instantly and are more responsive to changes in emotional states than peripheral neurophysiological signals. As a result, EEG readings may disclose crucial aspects of a person’s emotional states. The signals generated by electromyography (EMG) are gaining prominence in both clinical and biological settings. Differentiating between neuromuscular illnesses requires a reliable method of detection, processing, and classification of EMG data. This study investigates potential deep learning applications by constructing a framework to improve the prediction of cardiac-related diseases using electrocardiogram (ECG) data, furnishing an algorithmic model for sentiment classification utilizing EEG data, and forecasting neuromuscular disease classification utilizing EMG signals.

Keywords: Electrocardiography (ECG), electroencephalography (EEG), electromyographic (EMG), deeplearning techniques, prediction, heart attack, emotion recognition, neuromuscular disease, R-CNN

DOI: 10.3233/JIFS-230399

Journal: Journal of Intelligent & Fuzzy Systems, vol. 44, no. 6, pp. 9769-9782, 2023