Deformation corrected compressed sensing (DC-CS): a novel framework for accelerated dynamic MRI

TL;DR

This paper introduces a deformation corrected compressed sensing framework for accelerated dynamic MRI, enabling robust image reconstruction from undersampled data by handling various sparsity priors and motion correction.

Contribution

The novel DC-CS framework generalizes to multiple sparsity priors and employs variable splitting for efficient optimization, extending applications beyond cardiac cine to contrast-enhanced MRI.

Findings

Robust reconstruction with reduced motion artifacts

Effective handling of multiple sparsity priors

Applicable to diverse dynamic MRI scenarios

Abstract

We propose a novel deformation corrected compressed sensing (DC-CS) framework to recover dynamic magnetic resonance images from undersampled measurements. We introduce a generalized formulation that is capable of handling a wide class of sparsity/compactness priors on the deformation corrected dynamic signal. In this work, we consider example compactness priors such as sparsity in temporal Fourier domain, sparsity in temporal finite difference domain, and nuclear norm penalty to exploit low rank structure. Using variable splitting, we decouple the complex optimization problem to simpler and well understood sub problems; the resulting algorithm alternates between simple steps of shrinkage based denoising, deformable registration, and a quadratic optimization step. Additionally, we employ efficient continuation strategies to minimize the risk of convergence to local minima. The proposed formulation contrasts with existing DC-CS schemes that are customized for free breathing cardiac cine applications, and other schemes that rely on fully sampled reference frames or navigator signals to estimate the deformation parameters. The efficient decoupling enabled by the proposed scheme allows its application to a wide range of applications including contrast enhanced dynamic MRI. Through experiments on numerical phantom and in vivo myocardial perfusion MRI datasets, we demonstrate the utility of the proposed DC-CS scheme in providing robust reconstructions with reduced motion artifacts over classical compressed sensing schemes that utilize the compact priors on the original deformation un-corrected signal.