A Brief Overview of Optimization-Based Algorithms for MRI Reconstruction Using Deep Learning

TL;DR

This paper surveys optimization-based deep learning approaches to reconstruct MR images from undersampled k-space data, framing the problem within a learnable optimization (LOA) paradigm. It organizes methods around unrolled iterations of classical optimizers—gradient descent, proximal gradient, ADMM, and primal-dual methods—and discusses diffusion-model augmentations for robustness, including multi-coil and quantitative MRI. Key contributions include a consolidated view of how learnable proximal operators, data-consistency steps, and modality fusion can achieve accurate reconstructions while preserving interpretability, as well as highlighting diffusion-based techniques for handling distribution shifts. The work underscores the practical potential of LOA-based MRI reconstruction to improve speed, reliability, and clinical applicability, and points to future directions such as self-supervised training and transfer learning to further enhance performance and adaptability.

Abstract

Magnetic resonance imaging (MRI) is renowned for its exceptional soft tissue contrast and high spatial resolution, making it a pivotal tool in medical imaging. The integration of deep learning algorithms offers significant potential for optimizing MRI reconstruction processes. Despite the growing body of research in this area, a comprehensive survey of optimization-based deep learning models tailored for MRI reconstruction has yet to be conducted. This review addresses this gap by presenting a thorough examination of the latest optimization-based algorithms in deep learning specifically designed for MRI reconstruction. The goal of this paper is to provide researchers with a detailed understanding of these advancements, facilitating further innovation and application within the MRI community.