A video-based deep-learning system was trained to understand the spectrum of human cardiovascular disease by the self-supervised method of contrastive learning, using pairs of cardiac MRI scans and their corresponding text reports that are generated as part of routine clinical practice.
This is a preview of subscription content, access via your institution
Access options
Access Nature and 54 other Nature Portfolio journals
Get Nature+, our best-value online-access subscription
$32.99 / 30 days
cancel any time
Subscribe to this journal
Receive 12 digital issues and online access to articles
$119.00 per year
only $9.92 per issue
Buy this article
- Purchase on SpringerLink
- Instant access to the full article PDF.
USD 39.95
Prices may be subject to local taxes which are calculated during checkout
References
Moor, M. et al. Foundation models for generalist medical artificial intelligence. Nature 616, 259–265 (2023). A review article that discusses foundation models in medical artificial intelligence.
Zhang, Y., Jiang, H., Miura, Y., Manning, C. D. & Langlotz, C. P. Contrastive learning of medical visual representations from paired images and text. in Proc. Mach. Learn. Res. Vol. 182, 1–24 (PMLR, 2022). A seminal paper on the first successful attempt at asymmetrical contrastive learning for medical imaging.
Shad, R., Cunningham, J. P., Ashley, E. A., Langlotz, C. P. & Hiesinger, W. Designing clinically translatable artificial intelligence systems for high-dimensional medical imaging. Nat. Mach. Intell. 3, 929–935 (2021). A review article that discusses the challenges of deep-learning techniques with multi-dimensional imaging data.
Lipkova, J. et al. Deep learning-enabled assessment of cardiac allograft rejection from endomyocardial biopsies. Nat. Med. 28, 575–582 (2022). This paper uses multi-instance learning on pathology slides to achieve clinical-grade diagnostic performance at detecting organ rejection in recipients of heart transplant.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
This is a summary of: Shad, R. et al. A generalizable deep learning system for cardiac MRI. Nat. Biomed. Eng. https://doi.org/10.1038/s41551-026-01637-3 (2026).
Rights and permissions
About this article
Cite this article
Building foundation models for cardiac MRI. Nat. Biomed. Eng (2026). https://doi.org/10.1038/s41551-026-01638-2
Published:
Version of record:
DOI: https://doi.org/10.1038/s41551-026-01638-2
