SR-Net: A sequence offset fusion net and refine net for undersampled multislice MR image reconstruction. (April 2021)

• Record Type: Journal Article
• Authors: Xiao, Zhiyong; Du, Nianmao; Liu, Jianjun; Zhang, Weidong
• Is Part Of: Computer methods and programs in biomedicine. Volume 202(2021)
• Publication Date: 2021-04
• Subjects: Magnetic resonance imaging (MRI) -- Deep learning -- Deformable convolution -- Image reconstruction; Medicine -- Computer programs -- Periodicals; Biology -- Computer programs -- Periodicals; Computers -- Periodicals; Medicine -- Periodicals; Médecine -- Logiciels -- Périodiques; Biologie -- Logiciels -- Périodiques; Biology -- Computer programs; Medicine -- Computer programs; Periodicals; Electronic journals; 610.28
• DOI: 10.1016/j.cmpb.2021.105997
• Languages: English
• ISSNs: 0169-2607

Abstract

Highlights: We proposed an iterative architecture with long skip connections that can jointly exploit the correlations in a single slice and among slices. We built a bi-directional convolution subnet termed Slice Offset Fusion Net for correlations among slices. We used deformable convolutions to build a novel feature extractor that can help one slice extract useful information from its neighbor slices. We performed comprehensive experiments on two public raw MRI datasets with newly developed deep learning based methods and achieved competitive results in both quantitive metrics and visualization results. Abstract: Background and objective: The study of deep learning-based fast magnetic resonance imaging (MRI) reconstruction methods has become popular in recent years. However, there is still a challenge when MRI results undersample large acceleration factors. The objective of this study was to improve the reconstruction quality of undersampled MR images by exploring data redundancy among slices. Methods: There are two aspects of redundancy in multislice MR images including correlations inside a single slice and correlations among slices. Thus, we built two subnets for the two kinds of redundancy. For correlations among slices, we built a bidirectional recurrent convolutional neural network, named Sequence Offset Fusion Net (S-Net). In S-Net, we used a deformable convolution module to construct a neighbor slice feature extractor. For the correlation inside a single slice, we built a Refine Net (R-Net), which has 5 layers of 2D convolutions. In addition, we used a data consistency (DC) operation to maintain data fidelity in k-space. Finally, we treated the reconstruction task as a dealiasing problem in the image domain, and S-Net and R-Net are applied alternately and iteratively to generate the final reconstructions. Results: The proposed algorithm was evaluated using two online public MRI datasets. Compared with several state-of-the-art methods, the proposed method achieved better reconstruction results in terms of dealiasing and restoring tissue structure. Moreover, with over 14 slices per second reconstruction speed on 256x256 pixel images, the proposed method can meet the need for real-time processing. Conclusion: With spatial correlation among slices as additional prior information, the proposed method dramatically improves the reconstruction quality of undersampled MR images.