A Foundation Transformer Model with Self-Supervised Learning for ECG-Based Assessment of Cardiac and Coronary Function
This versatile ECG foundation model demonstrates that SSL pretraining enhances diagnostic accuracy and generalizability across diverse cardiac diagnostic applications. By enabling effective learning from limited labeled data, this approach supports AI development for complex but clinically critical tasks, such as detecting myocardial ischemia and coronary microvascular dysfunction, where high-quality labels are costly and scarce.
Authors
- Fares Alahdab (ORCID: https://orcid.org/0000-0001-5481-696X)
- Sascha N. Goonewardena (ORCID: https://orcid.org/0000-0002-3772-8818)
- Mouaz Husayn Al-Mallah (ORCID: https://orcid.org/0000-0003-2348-0484)
- Jennifer M. Renaud (ORCID: https://orcid.org/0000-0002-6408-3795)
- Jonathan B. Moody (ORCID: https://orcid.org/0000-0001-8177-0098)
- Alexis Poitrasson-Rivière (ORCID: https://orcid.org/0000-0002-6960-9389)
- Tomoe Hagio
- Venkatesh Locharla Murthy (ORCID: https://orcid.org/0000-0002-7901-1321)
- Michael D. Vanderver (ORCID: https://orcid.org/0009-0002-1309-9866)
- Edward P. Ficaro (ORCID: https://orcid.org/0000-0003-4803-2039)
Institutions
- Houston Methodist (US)
- VA Ann Arbor Healthcare System (US)
- Michigan Medicine (US)
- Institute for Medical Informatics and Biostatistics (CH)
- University of Missouri (US)
Publication Details
- Journal
- NEJM AI
- Published
- 2025-11-26
- DOI
- https://doi.org/10.1056/aioa2500164
- Citations
- 1
- Primary Topic
- ECG Monitoring and Analysis
- Type
- article
- Field-Weighted Citation Impact
- 1.39