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

arXiv cs.CV / 5/5/2026

📰 NewsDeveloper Stack & InfrastructureTools & Practical UsageModels & Research

共有:

Key Points

The paper proposes conditional score-based diffusion models (SBDMs) that take magnitude-only MR images and synthesize realistic phase maps compatible with the input magnitude.
It addresses the clinical reality that raw k-space data is often discarded, leaving only magnitude images and motivating training methods that can leverage magnitude-only registries.
The synthesized phase maps are used to generate large k-space datasets, which then train deep learning models for accelerated MRI reconstruction.
Experiments compare the resulting reconstruction model against baselines using smooth-phase assumptions, GAN-generated phase maps, and training with ground-truth k-space, and the SBDM-based approach shows improved quantitative metrics and reconstruction fidelity.
The key value is enabling more generalizable accelerated MRI reconstruction training from magnitude-only data while reducing reliance on limited or sensitive k-space datasets.

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.