Accelerated MR Elastography Using Learned Neural Network Representation

TL;DR

This paper introduces a deep learning approach for rapid, high-resolution MR elastography that reconstructs images from highly undersampled data using a learned neural network, improving quality over traditional methods.

Contribution

It presents a novel nonlinear neural network representation framework for MR elastography reconstruction from undersampled data, incorporating phase priors and self-supervised learning.

Findings

Superior image quality with noise and artifact suppression

Comparable stiffness estimation to fully sampled data

Effective from a single spiral arm per MRE repetition

Abstract

To develop a deep-learning method for achieving fast high-resolution MR elastography from highly undersampled data without the need of high-quality training dataset. We first framed the deep neural network representation as a nonlinear extension of the linear subspace model, then used it to represent and reconstruct MRE image repetitions from undersampled k-space data. The network weights were learned using a multi-level k-space consistent loss in a self-supervised manner. To further enhance reconstruction quality, phase-contrast specific magnitude and phase priors were incorporated, including the similarity of anatomical structures and smoothness of wave-induced harmonic displacement. Experiments were conducted using both 3D gradient-echo spiral and multi-slice spin-echo spiral MRE datasets. Compared to the conventional linear subspace-based approaches, the nonlinear network representation method was able to produce superior image reconstruction with suppressed noise and artifacts from a single in-plane spiral arm per MRE repetition (e.g., total R=10), yielding comparable stiffness estimation to the fully sampled data. This work demonstrated the feasibility of using deep network representations to model and reconstruct MRE images from highly-undersampled data, a nonlinear extension of the subspace-based approaches.