Score-based Generative Priors Guided Model-driven Network for MRI Reconstruction

TL;DR

This paper introduces a novel MRI reconstruction method that uses score-based priors and model-driven networks to improve image quality and reduce hallucination artifacts, especially with limited training data.

Contribution

The proposed workflow integrates pretrained score networks, denoising modules, and model-driven networks guided by denoised priors, eliminating the need for retraining and extensive hyperparameter tuning.

Findings

Outperforms existing methods in reducing hallucination artifacts.

Effective even with limited training data and sampling steps.

Robust high-quality MRI reconstructions achieved.

Abstract

Score matching with Langevin dynamics (SMLD) method has been successfully applied to accelerated MRI. However, the hyperparameters in the sampling process require subtle tuning, otherwise the results can be severely corrupted by hallucination artifacts, especially with out-of-distribution test data. To address the limitations, we proposed a novel workflow where naive SMLD samples serve as additional priors to guide model-driven network training. First, we adopted a pretrained score network to generate samples as preliminary guidance images (PGI), obviating the need for network retraining, parameter tuning and in-distribution test data. Although PGIs are corrupted by hallucination artifacts, we believe they can provide extra information through effective denoising steps to facilitate reconstruction. Therefore, we designed a denoising module (DM) in the second step to coarsely eliminate artifacts and noises from PGIs. The features are extracted from a score-based information extractor (SIE) and a cross-domain information extractor (CIE), which directly map to the noise patterns. Third, we designed a model-driven network guided by denoised PGIs (DGIs) to further recover fine details. DGIs are densely connected with intermediate reconstructions in each cascade to enrich the information and are periodically updated to provide more accurate guidance. Our experiments on different datasets reveal that despite the low average quality of PGIs, the proposed workflow can effectively extract valuable information to guide the network training, even with severely reduced training data and sampling steps. Our method outperforms other cutting-edge techniques by effectively mitigating hallucination artifacts, yielding robust and high-quality reconstruction results.