Classification of Electrocardiogram (ECG) Signals Based on Self-Supervised Learning
Academic Journal of Computing & Information Science · 2026 · 9 (2)
DOI: 10.25236/ajcis.2026.090201 ↗Kurz gefasst
Fachartikel in der Zeitschrift Academic Journal of Computing & Information Science, 2026.
Abstract (Originalsprache)
Electrocardiogram (ECG) analysis based on deep learning models has garnered significant research interest in recent years. Nevertheless, the performance of such models is often constrained by the limited availability of annotated ECG data.This paper proposes a novel self-supervised learning framework for ECG signal classification. Our method combines augmented contrastive learning with ECG-specific temporal augmentations (time truncation and random resized cropping). Experiments conducted on datasets (Cinc2020, Chapman, and Ribeiro) demonstrate that our approach achieves an average accuracy of 89.0%, an average AUC of 77.3%, and an F1-score of 63.3%. This represents improvements of 4.4%, 7.2%, and 3.8% in accuracy, AUC, and F1-score, respectively. When using only 60% of the labeled data, our method outperforms the fully supervised baseline by 7.7%. Ablation studies validate the effectiveness of our data augmentation strategies and contrastive learning design. Our approach offers a promising solution for label-efficient ECG analysis, with potential applications in clinical screening and remote monitoring systems.
Zitieren
APA
(2026). Classification of Electrocardiogram (ECG) Signals Based on Self-Supervised Learning. Academic Journal of Computing & Information Science, 9(2). https://doi.org/10.25236/ajcis.2026.090201
Vancouver
Classification of Electrocardiogram (ECG) Signals Based on Self-Supervised Learning. Academic Journal of Computing & Information Science. 2026;9(2). doi:10.25236/ajcis.2026.090201
BibTeX
@article{euvimed2026,
title = {Classification of Electrocardiogram (ECG) Signals Based on Self-Supervised Learning},
journal = {Academic Journal of Computing & Information Science},
year = {2026},
volume = {9},
number = {2},
doi = {10.25236/ajcis.2026.090201}
}
EUVIMED