Diffusion-QSM: Diffusion Model With Time-Travel and Resampling Refinement for Quantitative Susceptibility Mapping.

Journal: IEEE transactions on bio-medical engineering
Published Date: Apr 1, 2026

Abstract

OBJECTIVE: Quantitative susceptibility mapping (QSM) is a useful magnetic resonance imaging technique. We aim to propose a deep learning (DL)-based method for QSM reconstruction that is robust to data perturbations. METHODS: We developed Diffusion-QSM, a diffusion model-based method with a time-travel and resampling refinement module for high-quality QSM reconstruction. First, the diffusion prior is trained unconditionally on high-quality QSM images, without requiring explicit information about the measured tissue phase, thereby enhancing generalization performance. Subsequently, during inference, the physical constraints from the QSM forward model and measurement are integrated into the output of the diffusion model to guide the sampling process toward realistic image representations. In addition, a time-travel and resampling module is employed during the later sampling stage to refine the image quality, resulting in an improved reconstruction without significantly prolonging the time. RESULTS: Experimental results show that Diffusion-QSM outperforms traditional and unsupervised DL methods for QSM reconstruction using simulation, in vivo and ex vivo data and shows better generalization capability than supervised DL methods when processing out-of-distribution data. CONCLUSION: Diffusion-QSM successfully unifies data-driven diffusion priors and subject-specific physics constraints, enabling generalizable, high-quality QSM reconstruction under diverse perturbations, including image contrast, resolution and scan direction. SIGNIFICANCE: This work advances QSM reconstruction by bridging the generalization gap in deep learning. The excellent quality and generalization capability underscore its potential for various realistic applications.

Authors

  • Ming Zhang
    Heilongjiang Key Laboratory for Laboratory Animals and Comparative Medicine, College of Veterinary Medicine, Harbin 150030, China.
  • Chunlei Liu
    Department of Electrical Engineering and Computer Sciences, University of California, Berkeley, CA, USA.; Helen Wills Neuroscience Institute, University of California, Berkeley, CA, USA.
  • Yuyao Zhang
    School of Information and Science and Technology, ShanghaiTech University, Shanghai, China.
  • Hongjiang Wei
    Institute for Medical Imaging Technology, School of Biomedical Engineering, Shanghai Jiao Tong University, Shanghai, China. Electronic address: [email protected].

Keywords

No keywords available for this article.

External Resources

Popular Topics
Recent Journals