PROJECT TITLE :

Iterative Shrinkage Algorithm for Patch-Smoothness Regularized Medical Image Recovery

ABSTRACT:

We have a tendency to introduce a quick iterative shrinkage algorithm for patch-smoothness regularization of inverse problems in medical imaging. This approach is enabled by the reformulation of current non-local regularization schemes as an alternating algorithm to attenuate a global criterion. The proposed algorithm alternates between evaluating the denoised inter-patch variations by shrinkage and computing an image that is in line with the denoised inter-patch variations and measured knowledge. We tend to derive analytical shrinkage rules for several penalties that are relevant in non-native regularization. The redundancy in patch comparisons used to judge the shrinkage steps are exploited using convolution operations. The ensuing algorithm is observed to be significantly faster than current alternating non-local algorithms. The proposed theme is applicable to a large category of inverse problems together with deblurring, denoising, and Fourier inversion. The comparisons of the proposed theme with state-of-the-art regularization schemes in the context of recovering images from undersampled Fourier measurements demonstrate a considerable reduction in alias artifacts and preservation of edges.