Medical Diffusion: Denoising Diffusion Probabilistic Models for 3D Medical Image Generation
This addresses the need for privacy-preserving AI and data augmentation in medicine, though it is incremental by applying existing diffusion methods to medical data.
The authors tackled the problem of generating realistic 3D medical images using diffusion probabilistic models, showing they can synthesize high-quality MRI and CT images and improve breast segmentation performance from a dice score of 0.91 to 0.95 when data is scarce.
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.