Denoising diffusion probabilistic models for 3D medical image generation
Scientific Reports, vol. 13, pp. 7303
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 the synthesized images regarding "realistic image appearance", "anatomical correctness", and "consistency between slices". Furthermore, we demonstrate that synthetic images can be used in self-supervised pre-training and improve the performance of breast segmentation models when data is scarce (Dice scores, 0.91 [without synthetic data], 0.95 [with synthetic data]).
Authors 15
-
Affiliation as printed
Department of Diagnostic and Interventional Radiology, University Hospital Aachen, Aachen, Germany
-
Affiliation as printed
Department of Diagnostic and Interventional Radiology, University Hospital Aachen, Aachen, Germany
-
Affiliation as printed
Department of Diagnostic and Interventional Radiology, University Hospital Aachen, Aachen, Germany
-
Affiliation as printed
Physics of Molecular Imaging Systems, Experimental Molecular Imaging, RWTH Aachen University, Aachen, Germany
-
Affiliation as printed
Ocumeda GmbH, Munich, Germany
-
Affiliation as printed
Department of Diagnostic and Interventional Radiology, University Hospital Aachen, Aachen, Germany
-
Affiliation as printed
Department of Diagnostic and Interventional Radiology, University Hospital Aachen, Aachen, Germany
-
University Hospital Heidelberg
Affiliation as printed
Artificial Intelligence in Cardiovascular Medicine, University Hospital, Heidelberg, Germany
-
University of Würzburg · Universitätsklinikum Würzburg
Affiliation as printed
Department of Diagnostic and Interventional Radiology, University Hospital Würzburg, Würzburg, Germany
-
Johannes Gutenberg University Mainz
Affiliation as printed
Medical Clinic I, University of Mainz, Mainz, Germany
-
Affiliation as printed
Institute of Imaging and Computer Vision, RWTH Aachen, Aachen, Germany
-
Affiliation as printed
Department of Diagnostic and Interventional Radiology, University Hospital Aachen, Aachen, Germany
-
Affiliation as printed
Department of Diagnostic and Interventional Radiology, University Hospital Aachen, Aachen, Germany
-
University of Leeds · University Hospital Heidelberg · National Center for Tumor Diseases · Universitätsklinikum Aachen · Else Kröner Fresenius Center for Digital Health · Technische Universität Dresden
Affiliation as printed
Department of Medicine III, University Hospital Aachen, Aachen, Germany
Division of Pathology and Data Analytics, Leeds Institute of Medical Research at St James's, University of Leeds, Leeds, UK
Else Kroener Fresenius Center for Digital Health, Medical Faculty Carl Gustav Carus, Technical University Dresden, Dresden, Germany
Medical Oncology, National Center for Tumor Diseases (NCT), University Hospital Heidelberg, Heidelberg, Germany
-
Affiliation as printed
Department of Diagnostic and Interventional Radiology, University Hospital Aachen, Aachen, Germany. dtruhn@ukaachen.de
Department of Diagnostic and Interventional Radiology, University Hospital Aachen, Aachen, Germany
Cited by 341 stored of 343
Cited by patents worldwide 2 (Lens.org)
-
Image registration method and systemUS12555233B2 2026-02-17 Active