Dynamic Latent Variable Modeling for Improved Cardiac MRI Reconstruction
Motivation: The beating of the heart is predictable, and existing methods mainly focus on sparsity and low rank, ignore the predictability. Goal(s): Our goal is to improve the quality of dMRI reconstruction through predictability. Approach: Introduced a method to extract predictability latent vectors and reconstruct images based on it. Results: We used highly undersampled data for reconstruction and compared it with L+S,the experimental results indicate that we have achieved better reconstruction results than L+S. Impact: This work use predictability for dMRI reconstruction without the use of sparsity and low rank.It is possible to introduce a new perspective for the reconstruction of dynamic images.
Authors
- Zhihong Cui (ORCID: https://orcid.org/0000-0001-9729-5375)
- Haifeng Wang (ORCID: https://orcid.org/0000-0003-4229-3668)
- Jing Cheng (ORCID: https://orcid.org/0000-0001-9098-8048)
- Yanjie Zhu (ORCID: https://orcid.org/0000-0002-9131-689X)
- Sen Jia
- Shuo Zhou (ORCID: https://orcid.org/0000-0002-6475-3293)
- 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/10.58530/2024/1874
- Primary Topic
- Radiomics and Machine Learning in Medical Imaging
- Type
- article
- Field-Weighted Citation Impact
- 0.00