J-MoDL: Joint Model-Based Deep Learning for Optimized Sampling and Reconstruction

TL;DR

This paper presents J-MoDL, a joint deep learning framework that optimizes MRI sampling patterns and reconstruction networks simultaneously, leading to improved image quality in undersampled MRI data recovery.

Contribution

It introduces a continuous joint optimization method for sampling patterns and CNN parameters within a model-based deep learning MRI reconstruction scheme.

Findings

Joint optimization improves reconstruction quality across algorithms.

Continuous sampling pattern optimization enhances MRI scan efficiency.

Source code is publicly available for reproducibility.

Abstract

Modern MRI schemes, which rely on compressed sensing or deep learning algorithms to recover MRI data from undersampled multichannel Fourier measurements, are widely used to reduce scan time. The image quality of these approaches is heavily dependent on the sampling pattern. We introduce a continuous strategy to jointly optimize the sampling pattern and network parameters. We use a multichannel forward model, consisting of a non-uniform Fourier transform with continuously defined sampling locations, to realize the data consistency block within a model-based deep learning image reconstruction scheme. This approach facilitates the joint and continuous optimization of the sampling pattern and the CNN parameters to improve image quality. We observe that the joint optimization of the sampling patterns and the reconstruction module significantly improves the performance of most deep learning reconstruction algorithms. The source code of the proposed joint learning framework is available at https://github.com/hkaggarwal/J-MoDL.