Meeting Banner
Abstract #3806

Dictionary Based Reconstruction of Dynamic Complex MRI Data

Jose Caballero1, Anthony Price2, Daniel Rueckert1, Joseph V. Hajnal2

1Department of Computing, Imperial College London, London, United Kingdom; 2Division of Imaging Sciences and Biomedical Engineering Department, King's College London, London, United Kingdom

The reconstruction of MR data from undersampled observations has been studied as a solution to the acceleration of MR acquisition and shown to have great potential. Nonetheless, exploration of sparsity models has been somewhat limited, particularly in the case of dynamic MRI. Here we propose a combination of dictionary learning techniques and temporal gradient sparsity for the reconstruction of cardiac cine sequences. A comparison with an established method enforcing x-f support sparsity shows the benefits of carefully choosing a model. The technique presented is able to reconstruct the full complex data with an independent treatment of real and imaginary components.

Keywords

able accelerate accelerates achieve acquisition adapt aforementioned allows already around auxiliary background basis benchmarked best better biomedical caballero captured cardiac challenging cine coil college complex computing confirmed consider constraints convergence criterion dictionaries dictionary dimensions domain domains done dynamic easily enforcing essential even every examples feature find fixed frames full fully furthermore gradient gradients greatly ground imaginary impact importantly imposes impractical improve iteratively king kingdom learned learning like limited magazine magnitude mainly majority make many masks medicine modality model models nature noise open part particular parts patch patches peak performance powerful practice price problem processing produces propose proposed quality question real reconstruct reconstruction reconstructions recovery refer related representations represented revealed sampled sampling sciences seeks sense shortening solution solutions space sparse sparsely sparser sparsest sparsity speed square still subject subjects suboptimal subproblems suitable supplying support surpasses system systole taking task temporal though traditional train trained training transform transforms truth various vast wavelets white