EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

GANs for Medical Image Synthesis: An Empirical Study

Youssef Skandarani, Pierre-Marc Jodoin, Alain Lalande iD

Journal of Imaging · 16.03.2023 · 9 (3) · 69

DOI: 10.3390/jimaging9030069 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Imaging, online erschienen am 16.03.2023. Verfasst von Youssef Skandarani und 2 weiteren Autor:innen. Beteiligte Einrichtung: ImViA Laboratory, University of Bourgogne Franche-Comte, 21000 Dijon, France. Die Arbeit wurde bisher 276-mal zitiert (Crossref) und nennt 43 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Generative adversarial networks (GANs) have become increasingly powerful, generating mind-blowing photorealistic images that mimic the content of datasets they have been trained to replicate. One recurrent theme in medical imaging, is whether GANs can also be as effective at generating workable medical data, as they are for generating realistic RGB images. In this paper, we perform a multi-GAN and multi-application study, to gauge the benefits of GANs in medical imaging. We tested various GAN architectures, from basic DCGAN to more sophisticated style-based GANs, on three medical imaging modalities and organs, namely: cardiac cine-MRI, liver CT, and RGB retina images. GANs were trained on well-known and widely utilized datasets, from which their FID scores were computed, to measure the visual acuity of their generated images. We further tested their usefulness by measuring the segmentation accuracy of a U-Net trained on these generated images and the original data. The results reveal that GANs are far from being equal, as some are ill-suited for medical imaging applications, while others performed much better. The top-performing GANs are capable of generating realistic-looking medical images by FID standards, that can fool trained experts in a visual Turing test and comply to some metrics. However, segmentation results suggest that no GAN is capable of reproducing the full richness of medical datasets.

Schlagwörter

Stichwörter der Autor:innen

CTGANMRIadversarialheartliverretina

Autor:innen und Institutionen

  • ImViA Laboratory, University of Bourgogne Franche-Comte, 21000 Dijon, France.
  • CASIS Inc., 21800 Quetigny, France.
  • Department of Computer Science, University of Sherbrooke, Sherbrooke, QC J1K 2R1, Canada.
  • Department of Medical Imaging, University Hospital of Dijon, 21079 Dijon, France.

ORCID-Profile: Alain Lalande

Zitieren

APA

Skandarani, Y., Jodoin, P. M., & Lalande, A. (2023). GANs for Medical Image Synthesis: An Empirical Study. Journal of Imaging, 9(3), 69. https://doi.org/10.3390/jimaging9030069

Vancouver

Skandarani Y, Jodoin PM, Lalande A. GANs for Medical Image Synthesis: An Empirical Study. Journal of Imaging. 2023;9(3):69. doi:10.3390/jimaging9030069

BibTeX
@article{Skandarani2023,
  author = {Skandarani, Youssef and Jodoin, Pierre-Marc and Lalande, Alain},
  title = {GANs for Medical Image Synthesis: An Empirical Study},
  journal = {Journal of Imaging},
  year = {2023},
  volume = {9},
  number = {3},
  pages = {69},
  doi = {10.3390/jimaging9030069}
}
Kostenlos starten →