57 Translational evaluation of multimodal artificial intelligence for dermatology triage
Nuran Golbasi, Olivier Elemento, Veronica Rotemberg, Shari R Lipner · 2026
Read 57 Translational evaluation of multimodal artificial intelligence for dermatology triage by Nuran Golbasi, Olivier Elemento, Veronica Rotemberg, Shari R Lipner (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-13173214.txt.
Description
Nuran Golbasi and co-authors test six AI models for sorting skin lesions. Their 2026 study uses 200 lesions with biopsy proof, paired with photos and patient facts. The lesions were melanoma, other skin cancers, or harmless growths. Each model got the same prompt. It had to give a likely finding, a triage call, and its reasons.
All models caught most cancers, but they differed in how often they flagged harmless lesions. One model gave the best balance. Two others caught all cancers but flagged many harmless cases as urgent. Urgent calls matched skin doctors' biopsy choices in most cases. The team also checked fairness across groups and whether replies stayed the same when asked again.
Catalog record
| CREATOR | Nuran Golbasi, Olivier Elemento, Veronica Rotemberg, Shari R Lipner |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 2 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |