Temporal and spatial self supervised learning methods for electrocardiograms
Abstract The limited availability of labeled ECG data restricts the application of supervised deep learning methods in ECG detection. Although existing self-supervised learning approaches have been applied to ECG analysis, they are predominantly image-based, which limits their effectiveness. To addr...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2025-02-01
|
| Series: | Scientific Reports |
| Subjects: | |
| Online Access: | https://doi.org/10.1038/s41598-025-90084-2 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850191165857464320 |
|---|---|
| author | Wenping Chen Huibin Wang Lili Zhang Min Zhang |
| author_facet | Wenping Chen Huibin Wang Lili Zhang Min Zhang |
| author_sort | Wenping Chen |
| collection | DOAJ |
| description | Abstract The limited availability of labeled ECG data restricts the application of supervised deep learning methods in ECG detection. Although existing self-supervised learning approaches have been applied to ECG analysis, they are predominantly image-based, which limits their effectiveness. To address these limitations and provide novel insights, we propose a Temporal-Spatial Self-Supervised Learning (TSSL) method specifically designed for ECG detection. TSSL leverages the intrinsic temporal and spatial characteristics of ECG signals to enhance feature representation. Temporally, ECG signals retain consistent identity information over time, enabling the model to generate stable representations for the same individual across different time points while isolating representations of different leads to preserve their unique features. Spatially, ECG signals from various leads capture the heart’s activity from different perspectives, revealing both commonalities and distinct patterns. TSSL captures these correlations by maintaining consistency in the relationships between signals and their representations across different leads. Experimental results on the CPSC2018, Chapman, and PTB-XL databases demonstrate that TSSL introduces new capabilities by effectively utilizing temporal and spatial information, achieving superior performance compared to existing methods and approaching the performance of full-label training with only 10% of the labeled data. This highlights TSSL’s ability to provide deeper insights and enhanced feature extraction beyond mere performance improvements. We make our code publicly available on https://github.com/cwp9731/temporal-spatial-self-supervised-learning. |
| format | Article |
| id | doaj-art-7ed8a7ab2aab4dd2b30d82d1883c2bd0 |
| institution | OA Journals |
| issn | 2045-2322 |
| language | English |
| publishDate | 2025-02-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Scientific Reports |
| spelling | doaj-art-7ed8a7ab2aab4dd2b30d82d1883c2bd02025-08-20T02:15:00ZengNature PortfolioScientific Reports2045-23222025-02-0115111910.1038/s41598-025-90084-2Temporal and spatial self supervised learning methods for electrocardiogramsWenping Chen0Huibin Wang1Lili Zhang2Min Zhang3College of Information Science and Engineering, Hohai UniversityCollege of Computer Science and Software Engineering, Hohai UniversityCollege of Computer Science and Software Engineering, Hohai UniversityCollege of Information Science and Engineering, Hohai UniversityAbstract The limited availability of labeled ECG data restricts the application of supervised deep learning methods in ECG detection. Although existing self-supervised learning approaches have been applied to ECG analysis, they are predominantly image-based, which limits their effectiveness. To address these limitations and provide novel insights, we propose a Temporal-Spatial Self-Supervised Learning (TSSL) method specifically designed for ECG detection. TSSL leverages the intrinsic temporal and spatial characteristics of ECG signals to enhance feature representation. Temporally, ECG signals retain consistent identity information over time, enabling the model to generate stable representations for the same individual across different time points while isolating representations of different leads to preserve their unique features. Spatially, ECG signals from various leads capture the heart’s activity from different perspectives, revealing both commonalities and distinct patterns. TSSL captures these correlations by maintaining consistency in the relationships between signals and their representations across different leads. Experimental results on the CPSC2018, Chapman, and PTB-XL databases demonstrate that TSSL introduces new capabilities by effectively utilizing temporal and spatial information, achieving superior performance compared to existing methods and approaching the performance of full-label training with only 10% of the labeled data. This highlights TSSL’s ability to provide deeper insights and enhanced feature extraction beyond mere performance improvements. We make our code publicly available on https://github.com/cwp9731/temporal-spatial-self-supervised-learning.https://doi.org/10.1038/s41598-025-90084-2Deep learningSelf-supervised learningElectrocardiogramRepresentation extraction |
| spellingShingle | Wenping Chen Huibin Wang Lili Zhang Min Zhang Temporal and spatial self supervised learning methods for electrocardiograms Scientific Reports Deep learning Self-supervised learning Electrocardiogram Representation extraction |
| title | Temporal and spatial self supervised learning methods for electrocardiograms |
| title_full | Temporal and spatial self supervised learning methods for electrocardiograms |
| title_fullStr | Temporal and spatial self supervised learning methods for electrocardiograms |
| title_full_unstemmed | Temporal and spatial self supervised learning methods for electrocardiograms |
| title_short | Temporal and spatial self supervised learning methods for electrocardiograms |
| title_sort | temporal and spatial self supervised learning methods for electrocardiograms |
| topic | Deep learning Self-supervised learning Electrocardiogram Representation extraction |
| url | https://doi.org/10.1038/s41598-025-90084-2 |
| work_keys_str_mv | AT wenpingchen temporalandspatialselfsupervisedlearningmethodsforelectrocardiograms AT huibinwang temporalandspatialselfsupervisedlearningmethodsforelectrocardiograms AT lilizhang temporalandspatialselfsupervisedlearningmethodsforelectrocardiograms AT minzhang temporalandspatialselfsupervisedlearningmethodsforelectrocardiograms |