Medical Diffusion: Denoising Diffusion Probabilistic Models for 3D Medical Image Generation

TL;DR

This paper demonstrates that diffusion probabilistic models can generate high-quality 3D medical images like MRI and CT scans, improving data augmentation and model training in medical imaging.

Contribution

It introduces the application of diffusion probabilistic models to 3D medical image synthesis, showing their effectiveness and potential for enhancing medical AI tasks.

Findings

Synthetic images received high ratings for realism and anatomical correctness.

Using synthetic data improved breast segmentation accuracy in limited data scenarios.

Quantitative metrics confirmed the high quality of generated medical images.

Abstract

Recent advances in computer vision have shown promising results in image generation. Diffusion probabilistic models in particular have generated realistic images from textual input, as demonstrated by DALL-E 2, Imagen and Stable Diffusion. However, their use in medicine, where image data typically comprises three-dimensional volumes, has not been systematically evaluated. Synthetic images may play a crucial role in privacy preserving artificial intelligence and can also be used to augment small datasets. Here we show that diffusion probabilistic models can synthesize high quality medical imaging data, which we show for Magnetic Resonance Images (MRI) and Computed Tomography (CT) images. We provide quantitative measurements of their performance through a reader study with two medical experts who rated the quality of the synthesized images in three categories: Realistic image appearance, anatomical correctness and consistency between slices. Furthermore, we demonstrate that synthetic images can be used in a self-supervised pre-training and improve the performance of breast segmentation models when data is scarce (dice score 0.91 vs. 0.95 without vs. with synthetic data). The code is publicly available on GitHub: https://github.com/FirasGit/medicaldiffusion.