Article externe

AI skin cancer detection tools are getting better – but only for people with light skin

<name>Mohamed Akrout, Assistant Professor of Electrical Engineering and Computer Science, University of Tennessee</name> <foaf:homepage rdf:resource="https://theconversation.com/profiles/mohamed-akrout-2605973"/>· 3 septembre 2026

Melanoma and other skin conditions can look different in people with darker versus lighter skin – but image databases include many more examples from people with lighter skin. SeventyFour/iStock via Getty Images Imagine you’re getting out of the shower one morning and you notice a mole on your thig…

Résumé

AI skin cancer tools

New artificial intelligence (AI) tools are being developed to help detect skin cancer, including melanoma, a serious form of skin cancer.

These tools include smartphone apps for home use and software for doctors.

They analyze images of moles or skin lesions to identify potential signs of cancer.

AI models work by learning patterns from large datasets of medical images, associating visual features with specific diseases.

The goal is to provide accessible medical expertise, especially in areas with few dermatologists, potentially saving lives by enabling early detection.

However, current AI tools have a significant limitation: they perform well on light skin but struggle with darker skin tones.

Lire l'analyse complète en 6 points →

Lire l'article