Knowledge-driven deep learning for fast MR imaging: undersampled MR image reconstruction from supervised to un-supervised learning

TL;DR

This paper reviews how deep learning integrates domain knowledge with data-driven methods to improve fast MR imaging, highlighting challenges, solutions, and the evolution from supervised to unsupervised learning.

Contribution

It provides a comprehensive overview of knowledge-driven deep learning techniques for MR image reconstruction, emphasizing the transition from supervised to unsupervised approaches and discussing future directions.

Findings

Shift from supervised to unsupervised learning in MR reconstruction

Challenges in integrating physics-based models with deep learning

Discussion on MR vendors' adoption of DL techniques

Abstract

Deep learning (DL) has emerged as a leading approach in accelerating MR imaging. It employs deep neural networks to extract knowledge from available datasets and then applies the trained networks to reconstruct accurate images from limited measurements. Unlike natural image restoration problems, MR imaging involves physics-based imaging processes, unique data properties, and diverse imaging tasks. This domain knowledge needs to be integrated with data-driven approaches. Our review will introduce the significant challenges faced by such knowledge-driven DL approaches in the context of fast MR imaging along with several notable solutions, which include learning neural networks and addressing different imaging application scenarios. The traits and trends of these techniques have also been given which have shifted from supervised learning to semi-supervised learning, and finally, to unsupervised learning methods. In addition, MR vendors' choices of DL reconstruction have been provided along with some discussions on open questions and future directions, which are critical for the reliable imaging systems.