Data-Consistent Non-Cartesian Deep Subspace Learning for Efficient Dynamic MR Image Reconstruction

TL;DR

This paper introduces a data-consistent deep learning framework for non-Cartesian dynamic MRI reconstruction, significantly improving speed and accuracy over traditional iterative methods by developing four novel formulations and integrating them with U-Net models.

Contribution

It presents the first data-consistent deep learning approach specifically designed for non-Cartesian subspace MRI reconstruction, introducing four new formulations for enhanced performance.

Findings

Significant accuracy improvement over non-DC U-Net models

Reconstruction speed surpasses traditional iterative methods

Effective for cardiac MR multitasking imaging

Abstract

Non-Cartesian sampling with subspace-constrained image reconstruction is a popular approach to dynamic MRI, but slow iterative reconstruction limits its clinical application. Data-consistent (DC) deep learning can accelerate reconstruction with good image quality, but has not been formulated for non-Cartesian subspace imaging. In this study, we propose a DC non-Cartesian deep subspace learning framework for fast, accurate dynamic MR image reconstruction. Four novel DC formulations are developed and evaluated: two gradient decent approaches, a directly solved approach, and a conjugate gradient approach. We applied a U-Net model with and without DC layers to reconstruct T1-weighted images for cardiac MR Multitasking (an advanced multidimensional imaging method), comparing our results to the iteratively reconstructed reference. Experimental results show that the proposed framework significantly improves reconstruction accuracy over the U-Net model without DC, while significantly accelerating the reconstruction over conventional iterative reconstruction.