Distillation-Driven Diffusion Model for Multi-Scale MRI Super-Resolution: Make 1.5T MRI Great Again

2025-01-30

AI Summary

To address the trade-off between low resolution in 1.5T MRI and the high cost and limited clinical accessibility of 7T MRI, this paper proposes the first distillation-driven multi-scale diffusion super-resolution framework for 1.5T-to-7T-like image enhancement. Methodologically: (i) it incorporates 7T-specific physical priors—namely gradient nonlinearity and bias field correction—directly into the diffusion process for the first time; (ii) it introduces a progressive knowledge distillation strategy to construct an efficient lightweight student model; and (iii) it enables generalizable inference across arbitrary input resolutions without retraining. Evaluated on clinical MGH data, the student model reduces parameter count by 62% while incurring only a 0.18 dB PSNR degradation compared to the teacher, significantly improving lesion visibility and achieving state-of-the-art performance.

Technology Category

• Machine Learning: Multi-instance/Multi-view Learning
• Computer Vision: Diffusion Models for Vision
• Search and Optimization: Learning to Search

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Responsible Web: Data and user privacy-enhancing technologies for the Web

Abstract

Magnetic Resonance Imaging (MRI) offers critical insights into microstructural details, however, the spatial resolution of standard 1.5T imaging systems is often limited. In contrast, 7T MRI provides significantly enhanced spatial resolution, enabling finer visualization of anatomical structures. Though this, the high cost and limited availability of 7T MRI hinder its widespread use in clinical settings. To address this challenge, a novel Super-Resolution (SR) model is proposed to generate 7T-like MRI from standard 1.5T MRI scans. Our approach leverages a diffusion-based architecture, incorporating gradient nonlinearity correction and bias field correction data from 7T imaging as guidance. Moreover, to improve deployability, a progressive distillation strategy is introduced. Specifically, the student model refines the 7T SR task with steps, leveraging feature maps from the inference phase of the teacher model as guidance, aiming to allow the student model to achieve progressively 7T SR performance with a smaller, deployable model size. Experimental results demonstrate that our baseline teacher model achieves state-of-the-art SR performance. The student model, while lightweight, sacrifices minimal performance. Furthermore, the student model is capable of accepting MRI inputs at varying resolutions without the need for retraining, significantly further enhancing deployment flexibility. The clinical relevance of our proposed method is validated using clinical data from Massachusetts General Hospital. Our code is available at https://github.com/ZWang78/SR.

Problem

• MRI Image Enhancement
• 1.5T to 7T Resolution
• Medical Imaging Quality

Innovation

• Super-Resolution
• Teacher-Student Strategy
• 1.5T to 7T MRI Enhancement