Cape Town - 2026 ISMRM-ISMRT Annual Meeting and Exhibition
9 May 2026 – 14 May 2026 · Cape Town, South Africa
401-02-006 ISMRM Abstract

Foundation model for cardiac perfusion MRI enables 10-fold reduction in labeled dataset size for deep-learning analysis

Primary: Analysis Methods - Segmentation and Detection
Secondary: Analysis Methods - Foundation Models
401-02-006 · Foundation Models · Tuesday, 12 May, 8:20 AM–10:10 AM · Hall 1A
Keywords: Cardiac MRI AI Foundation Model Self-supervised learning Medical Image Segmentation
Accepted
M. Berk Sahin 1,2, Zhuoan Li1,3, Khalid Youssef4, Arian M Sohi1, Dilek M Yalcinkaya1,2, Luis Zamudio1,3, Michael Elliott5, Venkateshwar Polsani6, Matthew Tong7, Dipan Shah8, Orlando Simonetti9, Abolfazl Hashemi10, Behzad Sharif3,11
1Lab for Translational Imaging for Microcirculation, Purdue University, West Lafayette, United States of America
2Elmore School of Electrical & Computer Engineering, Purdue University, West Lafayette, United States of America
3Weldon School of Biomedical Engineering, Purdue University, West Lafayette, United States of America
4Radiology & Imaging Sciences, Indiana University School of Medicine, Indianapolis, United States of America
5Sanger Heart & Vascular Institute, Atrium Health, Charlotte, United States of America
6Piedmont Heart Institute, Piedmont Atlanta Hospital, Atlanta, United States of America
7Division of Cardiovascular Medicine, The Ohio State University, Columbus, United States of America
8Houston Methodist DeBakey Heart and Vascular Center, Houston, United States of America
9Department of Radiology, The Ohio State University, Columbus, United States of America
10Elmore Family School of Electrical and Computer Engineering, Purdue University, West Lafayette, United States of America
11Division of Cardiology, Indiana University School of Medicine, Indianapolis, United States of America
Presenting Author: M. Berk Sahin

Synopsis

Motivation:
Goals:
Approach:
Results:
Full abstract & presentation

The full text, figures, and any recorded presentation for this abstract are not shown here. Log in if you are a member or registered attendee with access.

Full abstracts, figures, and presentations for Cape Town - 2026 ISMRM-ISMRT Annual Meeting and Exhibition are available to registered attendees. This content becomes freely available to the public roughly two years after the meeting.

To request or purchase access, contact the ISMRM Central Office at info@ismrm.org.

Log in

References

1. Hachamovitch et al. Imaging registries and single-center series. JACC: Cardiovasc. Imaging 2017;10(3):276-85.
2. Leiner T, Rueckert D, Suinesiaputra A, et al. Machine learning in cardiovascular magnetic resonance: basic concepts and applications. J Cardiovasc Magn Reson 2019;21:61. doi: 10.1186/s12968-019-0575-y [doi]
3. Yalcinkaya D, Youssef K, Heydari B, et al. Improved robustness for deep learning-based segmentation of multi-center myocardial perfusion cardiovascular MRI datasets using data-adaptive uncertainty–guided space-time analysis. J Cardiovasc Magn Reson 2024;26:101082. doi: 10.1016/j.jocmr.2024.101082 [doi]
4. Jacob A, Borgohain I, Chitiboi T, et al. Towards a CMR Foundation Model for Multi-Task Cardiac Image Analysis, J Cardiovasc Magn Reson 2025; In Press. doi:10.1016/j.jocmr.2025.101967 [doi]
5. Dong, Haoyu, et al. MRI-CORE: A Foundation Model for Magnetic Resonance Imaging. arXiv preprint arXiv:2506.12186 (2025).
6. Fu, Yunguan, et al. CineMA: A Foundation Model for Cine Cardiac MRI. arXiv preprint arXiv:2506.00679 (2025).
7. He, Kaiming, et al. Masked autoencoders are scalable vision learners. Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. 2022. doi: 10.48550/arXiv.2111.06377 [doi]
8. Tong MS et al. The Society for Cardiovascular Magnetic Resonance Registry at 150,000. J Cardiovasc Magn Reson 2024;26(2):101055. doi: 10.1016/j.jocmr.2024.1010556 [doi]
9. Youssef K, Heydari B, Zamudio Rivero L, et al. A Patch-Wise Deep Learning Approach for Myocardial Blood Flow Quantification with Robustness to Noise and Nonrigid Motion. IEEE Proc. Eng Med Biol Soc 2021:4051-4057. doi: 10.1109/EMBC46164.2021.9629630 [doi]

Cite this abstract

http://echo.ismrm.org/p/ISMRM2026/401-02-006