EUVIMEDEUVIMED entdecken
Crossref

AI ‐generated dermatologic images show deficient skin tone diversity and poor diagnostic accuracy: An experimental study

Lucie Joerg, Margaret Kabakova, Jennifer Y Wang, Evan Austin, Marc Cohen, Alana Kurtti, Jared Jagdeo iD

Journal of the European Academy of Dermatology and Venereology · 2025-12 · 39 (12) · 2134-2141

DOI: 10.1111/jdv.20849 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of the European Academy of Dermatology and Venereology, online erschienen am 16.07.2025. Verfasst von Lucie Joerg und 6 weiteren Autor:innen. Beteiligte Einrichtung: Albany Medical College, Albany, New York, USA. Hauptthemen laut MEDLINE: Skin Diseases, Artificial Intelligence, Skin Pigmentation, Photography. Die Arbeit wurde bisher 30-mal zitiert (Crossref) und nennt 55 Literaturangaben.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Background

Generative AI models are increasingly used in dermatology, yet biases in training datasets may reduce diagnostic accuracy and perpetuate ethnic health disparities.

Objectives

To evaluate two key AI outputs: (1) skin tone representation and (2) diagnostic accuracy of generated dermatologic conditions.

Methods

Using the standard prompt ‘Generate a photo of a person with [skin condition],’ this cross‐sectional study investigated skin tone diversity and accuracy of four leading AI models—Adobe Firefly, ChatGPT‐4o, Midjourney and Stable Diffusion—across the 20 most common skin conditions. All images ( n = 4000) were evaluated for skin tone representation from June to July 2024. Two independent raters used the Fitzpatrick scale to assess skin tone diversity compared to U.S. Census demographics using χ 2 . Two blinded dermatology residents evaluated a randomized 200‐image subset for diagnostic accuracy. An inter‐rater kappa statistic was calculated to assess rater agreement.

Results

Across all generated images, 89.8% depicted light skin, and 10.2% depicted dark skin. Adobe Firefly demonstrated the highest alignment with U.S. demographic data, with a non‐significant chi‐square result (38.1% dark skin, χ 2 (1) = 0.320, p = 0.572), indicating no meaningful difference between its generated skin tone diversity and census demographics. ChatGPT‐4o, Midjourney and Stable Diffusion significantly underrepresented dark skin with Fitzpatrick scores of >IV (6.0%, 3.9% and 8.7% dark skin, respectively; all p < 0.001). Across all platforms, only 15% of images were identifiable by raters as the intended condition. Adobe Firefly had the lowest accuracy (0.94%), while ChatGPT‐4o, Midjourney and Stable Diffusion demonstrated higher but still suboptimal accuracy (22%, 12.2% and 22.5%, respectively).

Conclusions

The study highlights substantial deficiencies in the diversity and accuracy of AI‐generated dermatological images. AI programs may exacerbate cognitive bias and health inequity, suggesting the need for ethical AI guidelines and diverse datasets to improve disease diagnosis and dermatologic care.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Skin DiseasesArtificial IntelligenceSkin PigmentationPhotographyDermatologyHumansCross-Sectional StudiesFemaleMale

Stichwörter der Autor:innen

artificial intelligencedermatologydiagnostic accuracyracial disparitiesrepresentationskin of color

Autor:innen und Institutionen

  • Albany Medical College, Albany, New York, USA.
  • Department of Dermatology, State University of New York, Downstate Health Sciences University, Brooklyn, New York, USA.
  • Dermatology Service, Veterans Affairs New York Harbor Healthcare System - Brooklyn Campus, Brooklyn, New York, USA.

ORCID-Profile: Jared Jagdeo

Zitieren

APA

Joerg, L., Kabakova, M., Wang, J. Y., Austin, E., Cohen, M., Kurtti, A., & Jagdeo, J. (2025). AI ‐generated dermatologic images show deficient skin tone diversity and poor diagnostic accuracy: An experimental study. Journal of the European Academy of Dermatology and Venereology, 39(12), 2134-2141. https://doi.org/10.1111/jdv.20849

Vancouver

Joerg L, Kabakova M, Wang JY, Austin E, Cohen M, Kurtti A, et al. AI ‐generated dermatologic images show deficient skin tone diversity and poor diagnostic accuracy: An experimental study. Journal of the European Academy of Dermatology and Venereology. 2025;39(12):2134-2141. doi:10.1111/jdv.20849

BibTeX
@article{Joerg2025,
  author = {Joerg, Lucie and Kabakova, Margaret and Wang, Jennifer Y and Austin, Evan and Cohen, Marc and Kurtti, Alana and Jagdeo, Jared},
  title = {AI ‐generated dermatologic images show deficient skin tone diversity and poor diagnostic accuracy: An experimental study},
  journal = {Journal of the European Academy of Dermatology and Venereology},
  year = {2025},
  volume = {39},
  number = {12},
  pages = {2134--2141},
  doi = {10.1111/jdv.20849}
}
Kostenlos starten →