Multi-stage temporal and cross-view contrastive learning for self-supervised multi-lead ECG classification
Authors:
- Pengyao XuPXPengyao Xu
- Luyao LiLLLuyao Li
- Xiaoyun XieXXXiaoyun Xie
- Jiandong LvJLJiandong Lv
- Chong DiCDChong Di
- Xungang GuXGXungang Gu
- Yaoyao ChenYCYaoyao Chen
Open Access
- 1 December 2025
- journal article
- Published by Elsevier in Biomedical Signal Processing and Control
- Vol. 114
Abstract
No abstract availableFunding Information
- Qilu University of Technology (2023RCKY254)
- Shandong Academy of Sciences
- Natural Science Foundation of Shandong Province (ZR2022QF042)
- Natural Science Foundation of Shandong Province (ZR2021QE245)
- Ministry of Education of the People's Republic of China (2024PY038)
This publication has 23 references indexed in Scilit:
- Self-supervised contrastive learning achieves 12-lead ECG classificationBiomedical Signal Processing and Control, 2025
- ECG classification based on guided attention mechanismComputer Methods and Programs in Biomedicine, 2024
- Masked self-supervised ECG representation learning via multiview information bottleneckNeural Computing and Applications, 2024
- One-dimensional magnetotelluric parallel inversion using a ResNet1D-8 residual neural networkComputers & Geosciences, 2023
- Spatiotemporal self-supervised representation learning from multi-lead ECG signalsBiomedical Signal Processing and Control, 2023
- A lightweight 2-D CNN model with dual attention mechanism for heartbeat classificationApplied Intelligence, 2022
- A joint cross-dimensional contrastive learning framework for 12-lead ECGs and its heterogeneous deployment on SoCComputers in Biology and Medicine, 2022
- A simple self-supervised ECG representation learning method via manipulated temporal–spatial reverse detectionBiomedical Signal Processing and Control, 2022
- A multi-stage denoising framework for ambulatory ECG signal based on domain knowledge and motion artifact detectionFuture Generation Computer Systems, 2021
- Prediction of human protein subcellular localization using deep learningJournal of Parallel and Distributed Computing, 2018