Meeting Banner
Abstract #2621

Bayesian Compressive Sensing of Multishell HARDI for CSA-ODF Reconstruction

Julio Duarte-Carvajalino1, Christophe Lenglet1, Junqian Xu2, Essa S. Yacoub1, Kamil Ugurbil1, Steen Moeller1, Lawrence Carin3, Guillermo Sapiro3

1Center for Magnetic Resonance Research, University of Minnesota, Minneapolis, MN, United States; 2Mount Sinai School of Medicine, New York, NY, United States; 3Dept. of Electrical and Computer Engineering, Duke University, Durham, NC, United States

This work introduces a novel multi-task Bayesian compressive sensing approach for the direct and joint estimation of white matter fiber orientation distribution function and diffusion-weighted volumes from under-sampled HARDI data.

Keywords

acceleration accounts accurate acquisition acquisitions added additional advantages allows already among angular approaches benefit better blue blueprint bottom bundles challenge clinical clustering code coefficients color complementary compressive computed computer considered constant containing corrected coverage crossing currently dataset datasets dept diffusion directly distance distortions divergence duke eddy electrical employs enables engineering equal error especially estimating estimation euclidean fiber full future generated geometric gradient grants harmonics head histogram human improve improves in vivo indicate institutes intervention kinds kissing least library medicine middle model models motion mount need noise noising novel orientation orientations parametric partly phantom posterior previous produce project propose proposed protocol quality realistic received reconstructed reconstruction reduce reduction redundancy require resolution resolutions sampled schemes school sensing sets settings shell shells simultaneously since solid spatial spherical square staggered still support supported synthetic system table tables task tasks third toolkit tune tuned user usually various visual volumes