ECG-GraphNet: Advanced arrhythmia classification based on graph convolutional networks

ECG-GraphNet: Advanced arrhythmia classification based on graph convolutional networks

Background: Deep learning has significantly improved medical diagnostics, particularly in electrocardiogram (ECG) analysis, yet accurate classification of arrhythmias remains challenging. Objective: We propose Electrocardiogram Graph Convolutional Network (ECG-GraphNet), a graph convolutional networ...

Full description

Saved in:
Bibliographic Details
Main Authors: Myeonghun Lee, BS, Jiwoo Lim, MS, JinKook Kim, MS
Format: Article
Language:English
Published: Elsevier 2025-08-01
Series:Heart Rhythm O2
Subjects:
Electrocardiogram
Cardiovascular disease
Arrhythmia
Machine learning
Deep learning
Graph convolutional network
Online Access:http://www.sciencedirect.com/science/article/pii/S266650182500162X
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Background: Deep learning has significantly improved medical diagnostics, particularly in electrocardiogram (ECG) analysis, yet accurate classification of arrhythmias remains challenging. Objective: We propose Electrocardiogram Graph Convolutional Network (ECG-GraphNet), a graph convolutional network designed to classify arrhythmias into 3 types: normal (N), supraventricular ectopic (S), and ventricular ectopic (V) beats. Methods: ECG-GraphNet utilizes a novel graph representation of ECG data in which the P wave, QRS complex, and T wave are modeled as individual nodes. A unique QRS-centered weighted average pooling method is employed to enhance beat-specific feature extraction. We systematically explored various aspects including node features, edge definitions, a data augmentation method, and architecture configuration to determine the optimal model design. Experiments were conducted on 10-second ECG recordings from 328 patients using a single-lead device. Results: The optimized ECG-GraphNet achieved a Macro F1 score of 88.61% in 5-fold cross-validation. Scalability experiments further demonstrated its robustness, with Macro F1 scores of 85.21% and 87.03% across diverse ECG patterns and sizes. Conclusion: Our novel approach and comprehensive analysis underscore the potential advantages of ECG-GraphNet in clinical diagnosis and monitoring.
ISSN:2666-5018

Similar Items