Data-Consistent Non-Cartesian Deep Subspace Learning for Efficient Dynamic MR Image Reconstruction
This work addresses the clinical application bottleneck of slow reconstruction in dynamic MRI, representing an incremental improvement by adapting data-consistent deep learning to non-Cartesian subspace imaging.
The study tackled the slow iterative reconstruction problem in non-Cartesian dynamic MRI by proposing a data-consistent deep subspace learning framework, which improved reconstruction accuracy over a U-Net model without data consistency and accelerated reconstruction compared to conventional iterative methods.
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.