Cyclic Self-Supervised Diffusion for Ultra Low-field to High-field MRI Synthesis

2025-10-15

AI Summary

To address the low spatial resolution, poor signal-to-noise ratio, and insufficient anatomical fidelity of low-field MRI images, this paper proposes an unpaired cycle self-supervised diffusion framework for high-fidelity synthesis from ultra-low-field (≤0.05 T) to high-field (3 T) MRI. The method innovatively integrates a slice-level contrast-aware network with a local structural correction module, jointly optimizing anatomical consistency, cross-domain contrast matching, and fine-grained structural recovery during diffusion generation—bypassing conventional pixel-wise supervision. Cycle consistency constraints, self-reconstruction, and local masked perturbation recovery further enforce inter-slice structural alignment. Experiments demonstrate state-of-the-art performance: PSNR = 31.80 ± 2.70 dB, SSIM = 0.943 ± 0.102, and LPIPS = 0.0864 ± 0.0689. Crucially, anatomical error in white matter regions is reduced significantly—from 12.1% to 2.1%.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Machine Learning: Unsupervised & Self-Supervised Learning
• Natural Language Processing: Code Generation / Program Synthesis from Natural Language

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

Synthesizing high-quality images from low-field MRI holds significant potential. Low-field MRI is cheaper, more accessible, and safer, but suffers from low resolution and poor signal-to-noise ratio. This synthesis process can reduce reliance on costly acquisitions and expand data availability. However, synthesizing high-field MRI still suffers from a clinical fidelity gap. There is a need to preserve anatomical fidelity, enhance fine-grained structural details, and bridge domain gaps in image contrast. To address these issues, we propose a emph{cyclic self-supervised diffusion (CSS-Diff)} framework for high-field MRI synthesis from real low-field MRI data. Our core idea is to reformulate diffusion-based synthesis under a cycle-consistent constraint. It enforces anatomical preservation throughout the generative process rather than just relying on paired pixel-level supervision. The CSS-Diff framework further incorporates two novel processes. The slice-wise gap perception network aligns inter-slice inconsistencies via contrastive learning. The local structure correction network enhances local feature restoration through self-reconstruction of masked and perturbed patches. Extensive experiments on cross-field synthesis tasks demonstrate the effectiveness of our method, achieving state-of-the-art performance (e.g., 31.80 $pm$ 2.70 dB in PSNR, 0.943 $pm$ 0.102 in SSIM, and 0.0864 $pm$ 0.0689 in LPIPS). Beyond pixel-wise fidelity, our method also preserves fine-grained anatomical structures compared with the original low-field MRI (e.g., left cerebral white matter error drops from 12.1$%$ to 2.1$%$, cortex from 4.2$%$ to 3.7$%$). To conclude, our CSS-Diff can synthesize images that are both quantitatively reliable and anatomically consistent.

Problem

• Synthesizing high-field MRI from low-field scans
• Preserving anatomical fidelity in MRI synthesis
• Bridging domain gaps in image contrast

Innovation

• Cyclic self-supervised diffusion framework for MRI synthesis
• Slice-wise gap perception network aligns inter-slice inconsistencies
• Local structure correction network enhances feature restoration