PEARL: Cascaded Self-supervised Cross-fusion Learning For Parallel MRI Acceleration
Motivation: Supervised deep learning (SDL) has limitations due to data dependency, and self-supervised frameworks like DIP struggle with noise and artifacts. Goal(s): Introducing PEARL, a novel self-supervised accelerated parallel MRI approach. Approach: PEARL leverages joint deep decoders coupling with cross-fusion schemes based on multi-parameter priors to achieve enhanced reconstruction. Results: PEARL outperforms the existing methods, demonstrating notable improvement under highly accelerated acquisition. Impact: This study emphasizes the significance and potential of self-supervised learning in addressing critical MRI challenges.
Authors
- Haifeng Wang (ORCID: https://orcid.org/0000-0003-4229-3668)
- Zhuo‐Xu Cui (ORCID: https://orcid.org/0000-0001-9283-881X)
- Xiaomeng Yan (ORCID: https://orcid.org/0000-0002-6375-6310)
- Qingyong Zhu (ORCID: https://orcid.org/0000-0002-7817-3679)
- Hongwu Zeng (ORCID: https://orcid.org/0000-0002-7832-3702)
- Jing Cheng (ORCID: https://orcid.org/0000-0001-9098-8048)
- Chentao Cao (ORCID: https://orcid.org/0000-0002-3974-3413)
- Yanjie Zhu (ORCID: https://orcid.org/0000-0002-9131-689X)
- Yuanyuan Liu (ORCID: https://orcid.org/0000-0003-0465-3976)
- Bei Liu (ORCID: https://orcid.org/0000-0001-6294-4929)
- Dong Liang (ORCID: https://orcid.org/0000-0002-5582-4248)
- Yihang Zhou (ORCID: https://orcid.org/0009-0003-4852-6976)
Publication Details
- Journal
- Proceedings on CD-ROM - International Society for Magnetic Resonance in Medicine. Scientific Meeting and Exhibition/Proceedings of the International Society for Magnetic Resonance in Medicine, Scientific Meeting and Exhibition
- Published
- 2024-11-26
- DOI
- https://doi.org/10.58530/2024/1768
- Primary Topic
- Medical Imaging Techniques and Applications
- Type
- article
- Field-Weighted Citation Impact
- 0.00