Xiaoxin Li1,2, Xinjie Lou1, Junwei Yang3, Yong Chen4, and Dinggang Shen2,5
1College of Computer Science and Technology, Zhejiang University of Technology, Hangzhou, China, 2School of Biomedical Engineering, ShanghaiTech University, Shanghai, China, 3Department of Computer Science and Technology, University of Cambridge, Cambridge, United Kingdom, 4Case Western Reserve University, Cleveland, OH, United States, 5Shanghai United Imaging Intelligence Co., Ltd., Shanghai, China
1College of Computer Science and Technology, Zhejiang University of Technology, Hangzhou, China, 2School of Biomedical Engineering, ShanghaiTech University, Shanghai, China, 3Department of Computer Science and Technology, University of Cambridge, Cambridge, United Kingdom, 4Case Western Reserve University, Cleveland, OH, United States, 5Shanghai United Imaging Intelligence Co., Ltd., Shanghai, China
To accelerate multimodal Magnetic Resonance Imaging (MRI) acquisitions, we propose a deep Laplacian pyramid MRI reconstruction framework (LapMRI), which performs progressive upsampling while integrating multiscale prior of T1.
Figure 1: Schematic overview of the proposed LapMRI framework.
Figure 2: The inputs and outputs of LapMRI(D5C5, Λ=3) at three pyramid levels. For each pyramid level, the two images in the left column are the inputs, and the two images in the right column are the output and the respective ground-truth image, respectively. For visual understanding, the output of each pyramid level is displayed in the image domain, while the respective ground-truth image is framed with a red dotted box.