Lightweight Deep Learning Architecture for Multi-Lead ECG Arrhythmia Detection
Document Type
Article
Publication Date
2025
Abstract
Cardiovascular diseases are one of the top causes of death across the globe. Accurate detection and classification of cardiac arrhythmias from electrocardiogram (ECG) signals is essential for early diagnosing and treating cardiovascular diseases. This research introduces an innovative deep-learning architecture that integrates Convolutional Neural Networkswithachannelattention mechanism to classify five arrhythmia classes using both 2-lead and 12-lead ECG signals. The model’s performance was evaluated on the MIT-BIH and INCARTarrhythmia datasets, achieving accuracies of 99.18% and 99.48%, respectively, along with F1-scores of 99.18% and 99.48%. These high-performance metrics demonstrate the model’s effectiveness in distinguishing between normal and arrhythmic signals, as well as accurately identifying various arrhythmia types. The integration of attention mechanisms significantly enhanced the model’s capacity to concentrate on essential aspects of the ECG signals. This approach could improve the efficiency of healthcare systems and contribute to better patient outcomes.
Recommended Citation
yhiea, Nashwa Mohamed; Faculty of Science, Suez Canal Universirty; Faculty of Science, Suez Canal Universirt; and Professor at Graduate School of Information Science, University of Hyogo, "Lightweight Deep Learning Architecture for Multi-Lead ECG Arrhythmia Detection" (2025). Basic Science. 1.
https://buescholar.bue.edu.eg/basic_science/1