期刊
IEEE TRANSACTIONS ON MEDICAL IMAGING
卷 40, 期 12, 页码 3698-3710出版社
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/TMI.2021.3096218
关键词
Imaging; Image reconstruction; Learning systems; Transforms; Sparse matrices; Magnetic resonance imaging; Deep learning; Dynamic MR imaging; deep learning; compressed sensing; low-rank; model-based network
类别
资金
- National Key Research and Development Program of China [2017YFC0108802, 2020YFA0712200]
- National Natural Science Foundation of China [12026603, U1805261, 61771463, 81830056, 81971611, 61871373, 81729003]
- Key Field R&D Program of Guangdong Province [2018B030335001]
- Science and Technology Plan Program of Guangzhou [202007030002]
- Shenzhen Peacock Plan Team Program [KQTD20180413181834876]
- Innovation and Technology Commission of the Government of Hong Kong SAR [MRP/001/18X]
- Engineering Laboratory Program of Chinese Academy of Sciences [KFJ-PTXM-012]
- Strategic Priority Research Program of Chinese Academy of Sciences [XDC07040000, XDB25000000]
- Key Laboratory for Magnetic Resonance and Multimodality Imaging of Guangdong Province [2020B1212060051]
- China Postdoctoral Science Foundation [2020M682990, 2021M69331]
This paper proposes a method, named SLR-Net, which improves reconstruction results in dynamic MR imaging by exploring low-rank priors. Experimental results demonstrate that SLR-Net can further improve compressed sensing and deep learning methods in both single-coil and multi-coil scenarios, showing strong robustness to different undersampling patterns. The method shows capability and flexibility in real-time scenarios according to prospective reconstruction results on an open real-time dataset.
Deep learning methods have achieved attractive performance in dynamic MR cine imaging. However, most of these methods are driven only by the sparse prior of MR images, while the important low-rank (LR) prior of dynamic MR cine images is not explored, which may limit further improvements in dynamic MR reconstruction. In this paper, a learned singular value thresholding (Learned-SVT) operator is proposed to explore low-rank priors in dynamic MR imaging to obtain improved reconstruction results. In particular, we put forward a model-based unrolling sparse and low-rank network for dynamic MR imaging, dubbed as SLR-Net. SLR-Net is defined over a deep network flow graph, which is unrolled from the iterative procedures in the iterative shrinkage-thresholding algorithm (ISTA) for optimizing a sparse and LR-based dynamic MRI model. Experimental results on a single-coil scenario show that the proposed SLR-Net can further improve the state-of-the-art compressed sensing (CS) methods and sparsity-driven deep learning-based methods with strong robustness to different undersampling patterns, both qualitatively and quantitatively. Besides, SLR-Net has been extended to a multi-coil scenario, and achieved excellent reconstruction results compared with a sparsity-driven multi-coil deep learning-based method under a high acceleration. Prospective reconstruction results on an open real-time dataset further demonstrate the capability and flexibility of the proposed method on real-time scenarios.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据