STCP-Net: A few-shot learning approach based on spectral-temporal convolution and an improved prototypical network for ECG classification
PLOS One · 2026
DOI: 10.1371/journal.pone.0357954 ↗Kurz gefasst
Fachartikel in der Zeitschrift PLOS One, 2026. Verfasst von Jia Li und 8 weiteren Autor:innen. Beteiligte Einrichtung: School of Electronics and Information Engineering, Wuyi University, Jiangmen, China. Hauptthemen laut MEDLINE: Electrocardiography, Arrhythmias, Cardiac, Deep Learning. Die Arbeit nennt 42 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Guangdong Provincial Educational Research Project, The Third Division Tumxuk City Science and Technology Innovation Talent Program, China Postdoctoral Science Foundation.
Abstract (Originalsprache)
To address the heavy reliance of deep learning models on manually labeled data in ECG classification, we need to tackle two major limitations. First, sample sizes are often limited in clinical settings, making data annotation costly and impractical. Second, existing few-shot learning methods still fall short in two key areas: 1) Existing strategies pay insufficient attention to rare arrhythmia types. This makes it difficult to effectively identify rare clinical cases. 2) Single feature representation fails to fully integrate the global frequency, local time-frequency, and semantic information of ECG signals. This results in incomplete feature representation. To overcome these challenges, we propose a novel few-shot learning method for ECG classification. In terms of feature extraction, a spectral-temporal convolution feature extractor is designed to precisely focus on the key information of ECG by efficiently integrating the global frequency, local time-frequency features and semantic information of the signal. Furthermore, with regard to classification, an improved prototypical network based on cross-class bias diminishing is developed to enhance the recognition performance for rare arrhythmia type samples. The experimental results show that the proposed method achieves significant classification performance across multiple public datasets, especially on Chapman, it attains an average accuracy exceeding 90% in the 2-way 1-shot and 2-way 5-shot tasks.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Autor:innen und Institutionen
- School of Electronics and Information Engineering, Wuyi University, Jiangmen, China.
- School of Mathematics and Computational Science, Wuyi University, Jiangmen, China.
- Department of Cardiovascular Disease and Clinical Experimental Center, Jiangmen Central Hospital, Jiangmen, China.
- Department of Intensive Care Unit, Jiangmen Central Hospital, Jiangmen, China.
- Department of Electrocardiogram, Jiangmen Central Hospital, Jiangmen, China.
- Department of Rehabilitation, Shengjing Hospital of China Medical University, Shenyang, China.
Förderung
- Guangdong Provincial Educational Research Project
- The Third Division Tumxuk City Science and Technology Innovation Talent Program
- China Postdoctoral Science Foundation
Zitieren
APA
Li, J., Gan, D., Zhan, X., Yao, W., Zou, B., Zhang, B., Li, D., Li, D., & Li, L. (2026). STCP-Net: A few-shot learning approach based on spectral-temporal convolution and an improved prototypical network for ECG classification. PLOS One. https://doi.org/10.1371/journal.pone.0357954
Vancouver
Li J, Gan D, Zhan X, Yao W, Zou B, Zhang B, et al. STCP-Net: A few-shot learning approach based on spectral-temporal convolution and an improved prototypical network for ECG classification. PLOS One. 2026. doi:10.1371/journal.pone.0357954
BibTeX
@article{Li2026,
author = {Li, Jia and Gan, Dainan and Zhan, Xiaolin and Yao, Wei and Zou, Bin and Zhang, Bin and Li, Dong and Li, Dan and Li, Lin},
title = {STCP-Net: A few-shot learning approach based on spectral-temporal convolution and an improved prototypical network for ECG classification},
journal = {PLOS One},
year = {2026},
doi = {10.1371/journal.pone.0357954}
}
EUVIMED