Buchumschlag

Denoising diffusion probabilistic models for 3D medical image generation

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 th...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Khader, Firas (VerfasserIn) , Müller-Franzes, Gustav (VerfasserIn) , Tayebi Arasteh, Soroosh (VerfasserIn) , Han, Tianyu (VerfasserIn) , Haarburger, Christoph (VerfasserIn) , Schulze-Hagen, Maximilian (VerfasserIn) , Schad, Philipp (VerfasserIn) , Engelhardt, Sandy (VerfasserIn) , Baeßler, Bettina (VerfasserIn) , Foersch, Sebastian (VerfasserIn) , Stegmaier, Johannes (VerfasserIn) , Kuhl, Christiane (VerfasserIn) , Nebelung, Sven (VerfasserIn) , Kather, Jakob Nikolas (VerfasserIn) , Truhn, Daniel (VerfasserIn)
Dokumenttyp: Article (Journal)
Sprache:Englisch
Veröffentlicht: 2023
In: Scientific reports
Year: 2023, Jahrgang: 13, Pages: 1-12
ISSN:2045-2322
DOI:10.1038/s41598-023-34341-2
Online-Zugang:Verlag, kostenfrei, Volltext: https://doi.org/10.1038/s41598-023-34341-2
Verlag, kostenfrei, Volltext: https://www.nature.com/articles/s41598-023-34341-2
Volltext
Verfasserangaben:Firas Khader, Gustav Müller-Franzes, Soroosh Tayebi Arasteh, Tianyu Han, Christoph Haarburger, Maximilian Schulze-Hagen, Philipp Schad, Sandy Engelhardt, Bettina Baeßler, Sebastian Foersch, Johannes Stegmaier, Christiane Kuhl, Sven Nebelung, Jakob Nikolas Kather & Daniel Truhn
Beschreibung
Zusammenfassung: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]).
Beschreibung:Veröffentlicht: 05. Mai 2023
Gesehen am 04.08.2023
Beschreibung:Online Resource
ISSN:2045-2322
DOI:10.1038/s41598-023-34341-2

Ähnliche Einträge