MRI super-resolution in ten sampling steps using a diffusion bridge model

2026-08-09

AI Summary

This study addresses motion artifacts and limited resolution in MRI caused by prolonged scan times by proposing a Super-Resolution Diffusion Bridge Model (SR-DBM), which introduces diffusion bridge mechanisms into MRI super-resolution for the first time. The method formulates super-resolution as a stochastic transport process between low- and high-resolution image distributions, leveraging anatomical priors for initialization and incorporating Doob’s h-transform to constrain endpoint consistency. High-quality reconstructions are achieved with only ten steps of deterministic reverse sampling. Evaluated on 7T brain and prostate MRI datasets, SR-DBM significantly outperforms nine state-of-the-art methods (p < 0.05) in PSNR and SSIM metrics while achieving the lowest GMSD, demonstrating superior preservation of fine anatomical structures and lesions alongside markedly improved reconstruction efficiency and quality.

Abstract

Objective. MRI provides excellent soft-tissue contrast, but long acquisition times can cause patient discomfort and lead to motion artifacts, forcing a trade-off between spatial resolution and scan time. Diffusion-based super-resolution (SR) reconstructs high-resolution (HR) images from low-resolution (LR) inputs, but typically needs many sampling steps and initializes from a Gaussian prior ill-suited to image restoration. We developed an efficient diffusion framework that reconstructs HR MRI directly from LR data. Approach. We propose super-resolution diffusion bridge model (SR-DBM), a super-resolution diffusion bridge model that casts SR as a stochastic transport between the LR and HR image distributions. Through a Doob's h-transform of a mean-reverting stochastic differential equation, SR-DBM pins the process to the paired HR and LR images at its endpoints, initializing reconstruction from the measured anatomy rather than from Gaussian noise. The HR image is recovered by a deterministic reverse trajectory in which a network predicts the clean image at each of only ten sampling steps. We evaluated SR-DBM on ultra-high-field 7T brain T1 MP2RAGE maps and pelvic T2-weighted prostate images against nine comparison methods using PSNR, SSIM, GMSD, and LPIPS. Main results. SR-DBM attained the highest PSNR and SSIM and the lowest GMSD on both datasets (brain: 27.66+-1.52 dB, 0.96+-0.02, 7.96+-1.86$; prostate: 27.87+-2.29 dB, 0.80+-0.05, 8.38+- 1.44), with statistically significant gains over every comparison method (two-sided Wilcoxon signed-rank test with Holm correction, p<0.05). The strongest baseline, SR-EMamba, ranked second. Qualitatively, SR-DBM produced the smallest residual errors and best preserved fine structures and lesions.

Problem

• MRI super-resolution
• image reconstruction
• spatial resolution
• scan time
• motion artifacts

Innovation

• diffusion bridge
• MRI super-resolution
• Doob's h-transform
• stochastic transport
• few-step sampling