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

TL;DR

This paper introduces a diffusion-based super-resolution model that synthesizes 7T-like MRI images from 1.5T scans, employing a distillation strategy to create a lightweight, flexible, and high-performance tool validated on clinical data.

Contribution

It proposes a novel diffusion-driven MRI super-resolution framework with a progressive distillation approach for efficient, flexible, and high-quality 7T-like MRI synthesis from 1.5T scans.

Findings

State-of-the-art super-resolution performance achieved by the teacher model.

Lightweight student model maintains high performance with minimal sacrifice.

Student model accepts varying input resolutions without retraining.

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.