EFFICIENT ACCELERATED DYNAMIC MRI RECONSTRUCTION BASED ON ROBUST SEPARATION AND COMPRESSION OF LOW RANK AND SPARSE COMPONENTS

Abstract

The Low rank and Sparse (L+S) matrix decomposition model has been proposed in literature to reconstruct the accelerated dynamic MRI data. The limitations of L+S model include delicate separation and compression of the low-rank and sparse components from the acquired dynamic MRI data; also, the algorithm is computationally expensive. In this paper, Compressed Singular Value Decomposition (cSVD) is employed in L+S model for robust separation and P-Thresholding based compression of low rank and sparse components. The results show that the proposed method provides efficient reconstruction of accelerated dynamic MRI data in terms of quantifying parameters. Moreover, the proposed method possess highly parallel nature and demonstrate reduced in computation time using Graphic-Processing-Unit (GPU).