Adaptive Graph Attention Learning with Sparse Flow Optimization for Electrocardiogram-based Arrhythmia Classification

Read full text Summarise with VIRAT
Authors: J Hamsalatha, S Renukalatha
Cited by: 0
References: 27

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

Recently, arrhythmia has emerged as a major global health concern due to its rising incidence and potential to cause sudden cardiac events.Accurate classification of arrhythmia using Electrocardiogram (ECG) signals is essential for early diagnosis and clinical decision support.However, existing Deep Learning (DL) models suffer from noise sensitivity, baseline drift, class imbalance and limited generalization across diverse dataset, which reduced their ability in practical environments.To overcome these challenges, this study proposes a Graph Convolutional Network with Sparse Flow Induced Attention (GCN-SFIA) model for robust arrhythmia classification.The proposed GCN-SFIA integrated wavelet-based denoising to eliminate signal noise while preserving sharp morphological for accurate distortion-free cardiac signal processing.Subsequently, an adaptive graph connectivity is employed to capture interbeat dependencies effectively and SFIA is introduced to improve temporal-spatial feature extraction, thereby enabling robust discrimination of arrhythmia types.The results demonstrates that the proposed GCN-SFIA model attained better results in terms of accuracy, precision, recall and F1-score as 99.80% when compared to an existing 1Dimensional Convolutional Neural Network (1D CNN) model.

Fields of Study & Subjects

ECG Monitoring and Analysis Cardiac electrophysiology and arrhythmias Non-Invasive Vital Sign Monitoring

Publication details

Journal / source
International journal of intelligent engineering and systems
Volume
19
Issue
4
Pages
82–100
Publication date
23 March 2026
Language
EN