Taming Stable Diffusion for Computed Tomography Blind Super-Resolution

TL;DR

This paper introduces a novel method that adapts Stable Diffusion, a large-scale pre-trained diffusion model, for CT blind super-resolution, effectively enhancing image quality while reducing radiation exposure in medical imaging.

Contribution

It presents a new framework that combines realistic degradation modeling, text-guided supervision, and Stable Diffusion for improved CT super-resolution, addressing data scarcity and complex degradations.

Findings

Outperforms existing CT super-resolution methods

Achieves high-quality images at lower radiation doses

Demonstrates robustness across various degradation scenarios

Abstract

High-resolution computed tomography (CT) imaging is essential for medical diagnosis but requires increased radiation exposure, creating a critical trade-off between image quality and patient safety. While deep learning methods have shown promise in CT super-resolution, they face challenges with complex degradations and limited medical training data. Meanwhile, large-scale pre-trained diffusion models, particularly Stable Diffusion, have demonstrated remarkable capabilities in synthesizing fine details across various vision tasks. Motivated by this, we propose a novel framework that adapts Stable Diffusion for CT blind super-resolution. We employ a practical degradation model to synthesize realistic low-quality images and leverage a pre-trained vision-language model to generate corresponding descriptions. Subsequently, we perform super-resolution using Stable Diffusion with a specialized controlling strategy, conditioned on both low-resolution inputs and the generated text descriptions. Extensive experiments show that our method outperforms existing approaches, demonstrating its potential for achieving high-quality CT imaging at reduced radiation doses. Our code will be made publicly available.