articleScientific ReportsMay 5, 2023GOLD OA

Denoising diffusion probabilistic models for 3D medical image generation

FKFiras KhaderGMGustav Müller‐FranzesSTSoroosh Tayebi ArastehTHTianyu HanCHChristoph Haarburger

Universitätsklinikum Aachen · RWTH Aachen University · +7 more institutions

PubMed
Indexed incrossrefdoajpubmed

Abstract

Recent advances in computer vision have shown promising results in image generation. Diffusion probabilistic models have generated realistic images from textual input, as demonstrated by DALL-E 2, Imagen, and Stable Diffusion. However, their use in medicine, where imaging 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. We show that diffusion probabilistic models can synthesize high-quality medical data for magnetic resonance imaging (MRI) and computed tomography (CT). For quantitative evaluation, two radiologists rated the quality of…

Citation impact

292
total citations
FWCI
48.38
Percentile
100%
References
32
Citations per year

Authors

15

Topics & keywords

Topics
Keywords
  • Computer science
  • Probabilistic logic
  • Artificial intelligence
  • Correctness
  • Synthetic data
  • Medical imaging
  • Segmentation
  • Image quality
No related works found for this paper.

Funding