Title: ResNet-based deep learning approach for automated ECG arrhythmia recognition system

Authors: Soumen Ghosh; Satish Chander

Addresses: Birla Institute of Technology, Mesra, India ' Birla Institute of Technology, Mesra, India

Abstract: This study introduces a novel approach to classifying electrocardiogram (ECG) signals using deep learning techniques, specifically leveraging TensorFlow and the ResNet50 architecture. The research aims to develop a robust model for accurate ECG signal classification, which is vital for various medical diagnostics and healthcare applications. The methodology includes preprocessing the ECG dataset, constructing a deep neural network model based on ResNet50, and training it on the labelled data. The model's performance is thoroughly evaluated using metrics like accuracy, confusion matrix, classification report, F1-score, and ROC curve analysis. The results show a 92.71% accuracy in ECG signal classification, highlighting the effectiveness of the proposed approach. This research advances signal classification in healthcare, offering a promising tool for the automated analysis and interpretation of ECG signals, ultimately aiding healthcare professionals in the timely diagnosis and treatment of cardiovascular conditions.

Keywords: deep learning; signal classification; electrocardiogram; ECG; ResNet50; TensorFlow; medical diagnostics; healthcare; accuracy; performance evaluation; confusion matrix; classification report; F1-score; ROC curve analysis; automated analysis; cardiovascular conditions.

DOI: 10.1504/IJBRA.2025.149724

International Journal of Bioinformatics Research and Applications, 2025 Vol.21 No.5, pp.506 - 521

Received: 26 Apr 2024
Accepted: 05 Aug 2024
Published online: 11 Nov 2025 *

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