Title: Applications of deep learning algorithms in biomedical signal processing - pros and cons

Authors: Gyana Ranjan Patra; Saumendra Kumar Mohapatra; Mihir Narayan Mohanty

Addresses: Department of Electronics and Communication Engineering, ITER (FET), Siksha 'O' Anusandhan (Deemed to be University), Bhubaneswar, India ' Department of Computer Science and Engineering, ITER (FET), Siksha 'O' Anusandhan (Deemed to be University), Bhubaneswar, India ' Department of Electronics and Communication Engineering, ITER (FET), Siksha 'O' Anusandhan (Deemed to be University), Bhubaneswar, India

Abstract: The recently introduced deep neural networks are artificial neural networks that contain multiple hidden layers and can facilitate computational models to acquire data representation with multiple levels of abstraction, thereby showing consistently good results in the areas of speech/visual object recognition, object detection, and many more. In this paper, a study on different biomedical signals is carried out with a closed look on cardiac signal since cardiac failure gets reported a lot, and research on this signal necessitates a special attention. The authors have approached the ECG classification with deep learning model with the study of their pros and cons. A methodology for ECG classification using a 10-layer convolutional neural network has been proposed. Results have shown that the proposed model is capable of performing better than some of the other methods like support vector machine and wavelet transformation methods.

Keywords: electrocardiogram; ECG; ECG classification; deep learning; deep neural networks.

DOI: 10.1504/IJBM.2022.119560

International Journal of Biometrics, 2022 Vol.14 No.1, pp.98 - 124

Received: 11 Dec 2019
Accepted: 06 Oct 2020

Published online: 09 Dec 2021 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article