Phase-map synthesis from magnitude-only MR images using conditional score-based diffusion models with application in training of accelerated MRI reconstruction models

2026-05-01

AI Summary

This work addresses the limitation in clinical MRI where only magnitude images are routinely archived while raw k-space data—including phase information—are discarded, hindering the development of accelerated reconstruction models. To overcome this, the study proposes the first application of a conditional score-based diffusion model to synthesize realistic phase maps from single magnitude images, thereby reconstructing complete k-space data. By leveraging vast repositories of anonymized magnitude images, the method generates high-quality synthetic training data that significantly enhances the generalization capability of downstream reconstruction models. Experimental results demonstrate that models trained on the synthesized k-space data outperform baseline approaches—including those using smoothed phase or GAN-generated phase—in both quantitative metrics and image fidelity, effectively suppressing artifacts and hallucinated features.

Abstract

Accelerated magnetic resonance imaging (MRI) enabled by the training of deep learning (DL)-based image recon. models requires large and diverse raw k-space datasets. In most clinical MRI applications, due to storage and patient privacy concerns, raw k-space data is discarded and magnitude-only images are the only component saved. Consequently, a large portion of the DL-based MRI recon. literature has either relied on small training datasets or has used one of the few available open-source k-space datasets. At the same time, the growing number of anonymized magnitude-only image registries/databases motivates the development of techniques that can use them as training datasets for generalizable DL-based recon. models. Here we propose to address this challenge by employing a generative approach based on conditional score-based diffusion models (SBDMs): given a magnitude-only MR image, it synthesizes a phase map (in the image domain) that realistically corresponds to the magnitude-only image. We evaluate its generative capabilities in a downstream DL-based recon. task whereby a large k-space dataset is generated by combining the SBDM-synthesized phase-maps and the corresponding magnitude-only images, and this k-space dataset is then used to train a DL model for accelerated MRI recon. We compare the performance of the resulting DL model versus those trained according to (a) a naive approach that uses smooth phase, (b) a k-space training dataset generated using synthesized phase maps derived from a generative adversarial network, and (c) the ground truth k-space data. Our results suggest that the DL model trained from SBDM-synthesized k-space data outperforms the other approaches in terms of quantitative metrics as well as qualitatively observed recon. fidelity, i.e., whether the reconstructed images include erroneous or hallucinated features that could adversely impact diagnostic accuracy.

Problem

• accelerated MRI
• magnitude-only images
• k-space data
• deep learning reconstruction
• phase map synthesis

Innovation

• score-based diffusion models
• phase-map synthesis
• accelerated MRI reconstruction
• magnitude-only MRI
• k-space data generation