Yunsong Liu

TL;DR: In this paper, an iteration rebuilding method of a magnetic resonance image is proposed, which is simple in calculation, fast in operation speed and few in computer memory, and involves in the magnetic resonance imaging.

TL;DR: This paper attempts to address the problem of applying fast iterative soft-thresholding algorithm (FISTA) to tight frames based magnetic resonance image reconstruction by introducing the canonical dual frame, and proposes a new algorithm, called the projected FISTA (pF ISTA), which theoretically proves that pFISta converges to the minimum of a function with a balanced tight frame sparsity.

TL;DR: A novel implementation of the previous proximal alternating linearized minimization (PALM) algorithm is described, leading to a new algorithm named PALMNUT that combines the PALM together with Nesterov's momentum and a novel approach that relies on uncoupled coordinatewise step sizes derived from coordinatewise Lipschitz-like bounds.

TL;DR: In this article, the authors proposed a new algorithm for complex image reconstruction with separate regularization of the image magnitude and phase, named PALMNUT, which combines the proximal alternating linearized minimization (PALM) algorithm with Nesterov's momentum and relies on uncoupled coordinatewise step sizes derived from coordinatewise Lipschitz-like bounds.

TL;DR: A parallel architecture based on multicore processors is proposed to accelerate computations of PANO and results demonstrate that the acceleration factor approaches the number of CPU cores and overall PANO-based CS-MRI reconstruction can be accomplished in several seconds.