Meeting Banner
Abstract #2046

Compressive Diffusion MRI Part 2: Performance Evaluation Via Low-Rank Model

Hao Gao1, 2, Longchuan Li3, Xiaoping P. Hu3

1Department of Mathematics and Computer Science, Emory University, Atlanta, GA, United States; 2Department of Radiology and Imaging Sciences, Emory University, Atlanta, GA, United States; 3Department of Biomedical Engineering, Emory University and Georgia Institute of Technology, Atlanta, GA, United States

In another submitted abstract Compressive Diffusion MRI Part 1: Why Low-Rank?, we compared several sparsity models and found that the low-rank (LR) model is the most suitable for diffusion MRI. In this abstract we retrospectively explore compressive MRI in the context of diffusion MRI via LR. The results suggest that LR is able to accurately reconstruct the diffusion MR images from highly undersampled k-space, in terms of both the image quality and the angular differences in the principal and secondary diffusion orientations.

Abstract PDF

Keywords

able abstract accurately allow angular another autism biomedical biomedicine brain chosen compartment compressive computer connectivity context contrasts correcting diffusion distortion distortions eddy emerging encoding engineering errors evaluation example explore exploring extremely facilitate fold generated gold gray greatly hand healthy highly human improved institute integrity isotropic larger limitation long major mathematics matrix middle minute model models molecules newborns opposite orientation orientations part partial patient performance plotted potentially powerful principal process proportion quality quantitative radiology randomly rank reconstruct reconstructed reconstruction reduced reduction related relatively removal resolution resp respectively retrospectively science sciences secondary segmented sensitive separately several simulated since slice space sparsity spatial subject submitted subsequently suggest suitable susceptibility technology terms tissue utilizes various volume water white