Exploiting both intra-quadtree and inter-spatial structures for multi-contrast MRI

Multi-contrast magnetic resonance images are not only compressible but also share the same inter-spatial structure as they are scanned from the same anatomical cross section. In addition, the wavelet coefficients of a MR image naturally yield an intra-quadtree structure and has been used in compressed imaging. In this paper, we propose a new method to reconstruct multi-contrast MR images by exploiting their intra- and inter- structures simultaneously. Based on structured sparsity theory, it could further reduce the undersampled data for reconstruction or enhance the reconstruction quality. A new algorithm is proposed to efficiently solve this problem. Experiments demonstrate the superiority of the proposed algorithm over existing methods on multi-contrast MRI.