EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Skin Tone in Hyperspectral Imaging and Its Implications for Fairness in AI

Laurie S van de Weerd iD, Nick J van de Berg, L Lucia Rijstenberg, Ralf L O van de Laar, Helena C van Doorn, Heleen J van Beekhuizen

Journal of Biophotonics · 2026-04 · 19 (4)

DOI: 10.1002/jbio.70254 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Biophotonics, online erschienen am 31.03.2026. Verfasst von Laurie S van de Weerd und 5 weiteren Autor:innen. Beteiligte Einrichtung: Department of Gynaecological Oncology, Erasmus MC Cancer Institute, University Medical Center Rotterdam, Rotterdam, the Netherlands. Hauptthemen laut MEDLINE: Artificial Intelligence, Hyperspectral Imaging, Image Processing, Computer-Assisted, Skin. Die Arbeit nennt 34 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Artificial intelligence (AI) is increasingly applied in healthcare, but concerns remain about bias affecting under‐represented groups. We investigated whether skin tone is systematically encoded in hyperspectral imaging data and how this affects classifications. Images were collected from 45 healthy women of the upper leg skin and vulvar mucosal tissue. Skin tones were grouped using the individual typology angle scale. Physiological parameters (oxygen saturation, haemoglobin, water and near‐infrared indices) were compared across groups. Unsupervised and supervised classification models were evaluated. Skin tone values ranged from −0.7 to 75.8 (20 very light, 9 light, 9 intermediate, 7 tan and 2 brown). All physiological parameters differed significantly across groups ( p < 0.001). Unsupervised learning achieved 38.5% balanced accuracy, whereas supervised learning reached 71.4%, with high accuracies for tan (94.6%) and brown (95.0%) groups. Skin tone influences HSI data; it may act as a confounder in AI models, underscoring the need for diverse datasets to ensure equitable performance.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Artificial IntelligenceHyperspectral ImagingImage Processing, Computer-AssistedSkinSkin PigmentationHumans

Stichwörter der Autor:innen

artificial intelligence (AI)health equityoptical imagingskin pigmentation

Autor:innen und Institutionen

  • Department of Gynaecological Oncology, Erasmus MC Cancer Institute, University Medical Center Rotterdam, Rotterdam, the Netherlands.
  • Department of Pathology, Erasmus University Medical Center Rotterdam, Rotterdam, the Netherlands.

ORCID-Profile: Laurie S van de Weerd

Zitieren

APA

van de Weerd, L. S., van de Berg, N. J., Rijstenberg, L. L., van de Laar, R. L. O., van Doorn, H. C., & van Beekhuizen, H. J. (2026). Skin Tone in Hyperspectral Imaging and Its Implications for Fairness in AI. Journal of Biophotonics, 19(4). https://doi.org/10.1002/jbio.70254

Vancouver

van de Weerd LS, van de Berg NJ, Rijstenberg LL, van de Laar RLO, van Doorn HC, van Beekhuizen HJ. Skin Tone in Hyperspectral Imaging and Its Implications for Fairness in AI. Journal of Biophotonics. 2026;19(4). doi:10.1002/jbio.70254

BibTeX
@article{vandeWeerd2026,
  author = {van de Weerd, Laurie S and van de Berg, Nick J and Rijstenberg, L Lucia and van de Laar, Ralf L O and van Doorn, Helena C and van Beekhuizen, Heleen J},
  title = {Skin Tone in Hyperspectral Imaging and Its Implications for Fairness in AI},
  journal = {Journal of Biophotonics},
  year = {2026},
  volume = {19},
  number = {4},
  doi = {10.1002/jbio.70254}
}
Kostenlos starten →