A multi-dynamic low-rank deep image prior for three-dimensional real-time cardiovascular magnetic resonance imaging
ML-DIP enables high-quality 3D real-time CMR with acceleration factors exceeding 1000 by learning low-rank spatial and motion representations from undersampled data, without relying on external fully sampled training datasets.
Authors
- Zhenyu Bu
- Syed Murtaza Arshad (ORCID: https://orcid.org/0000-0001-8377-0710)
- Preethi Chandrasekaran (ORCID: https://orcid.org/0000-0002-1581-6008)
- Marc Vornehm (ORCID: https://orcid.org/0009-0004-5687-0563)
- Yingmin Liu (ORCID: https://orcid.org/0000-0001-5886-8636)
- Muhammad Ahmad Sultan (ORCID: https://orcid.org/0009-0007-0185-7400)
- Yuchi Han (ORCID: https://orcid.org/0000-0001-7582-1848)
- R. Badlishah Ahmad (ORCID: https://orcid.org/0000-0002-4862-2728)
- Chong Chen (ORCID: https://orcid.org/0000-0002-4965-5564)
Institutions
- Friedrich-Alexander-Universität Erlangen-Nürnberg (DE)
- Lung Institute (US)
- Siemens Healthineers (Germany) (DE)
- The Ohio State University (US)
Publication Details
- Journal
- Journal of Cardiovascular Magnetic Resonance
- Published
- 2025-11-24
- DOI
- https://doi.org/10.1016/j.jocmr.2025.102015
- Primary Topic
- Advanced MRI Techniques and Applications
- Type
- article
- Field-Weighted Citation Impact
- 0.00