A Motion Assessment Method for Fetal Brain MRI Based on 2D-SVD and 3D-CP Decomposition
Haoan Xu1, Wen Shi1,2, Jiwei Sun1, Cong Sun3, Guangbin Wang3,4, Yi Zhang1, and Dan Wu1
1Key Laboratory for Biomedical Engineering of Ministry of Education, Department of Biomedical Engineering, College of Biomedical Engineering & Instrument Science, Zhejiang University, Hangzhou, China, 2Department of Biomedical Engineering, Johns Hopkins University School of Medicine, Baltimore, MD, United States, 3Department of Radiology, Cheeloo College of Medicine, Shandong Provincial Hospital, Shandong University, Jinan, China, 4Department of Radiology, Shandong Provincial Hospital Affiliated to Shandong First Medical University, Jinan, China
Slice-to-volume registration and super-resolution reconstruction is commonly used to generate 3D volumes of the fetal brain from 2D stacks in multiple orientations. The current pipeline requires selecting the stack with minimal motion as a reference for registration. We proposed a motion assessment method that automatically determines the reference stack based on CANDECOMP/PARAFAC decomposition. This method is sensitive to motion across slices compared to other state-of-the-art methods. Combining motion assessment with the existing fetal brain MRI processing pipeline improved the reconstruction quality and the success rate.
This abstract and the presentation materials are available to members only;
a login is required.