K-space and Image Domain Collaborative Energy based Model for Parallel MRI Reconstruction

TL;DR

This paper introduces a hybrid deep energy-based model that collaboratively reconstructs MRI images directly from under-sampled k-space data, improving accuracy and stability over existing methods.

Contribution

It presents the first deep energy-based model that operates in both k-space and image domains for MRI reconstruction, enhancing robustness and flexibility.

Findings

Lower reconstruction error compared to state-of-the-art methods

More stable performance across various acceleration factors

Effective hybrid domain approach for MRI data estimation

Abstract

Decreasing magnetic resonance (MR) image acquisition times can potentially make MR examinations more accessible. Prior arts including the deep learning models have been devoted to solving the problem of long MRI imaging time. Recently, deep generative models have exhibited great potentials in algorithm robustness and usage flexibility. Nevertheless, none of existing schemes can be learned or employed to the k-space measurement directly. Furthermore, how do the deep generative models work well in hybrid domain is also worth being investigated. In this work, by taking advantage of the deep energy-based models, we propose a k-space and image domain collaborative generative model to comprehensively estimate the MR data from under-sampled measurement. Experimental comparisons with the state-of-the-arts demonstrated that the proposed hybrid method has less error in reconstruction accuracy and is more stable under different acceleration factors