A 10-year-old CNN outperforms frontier LLMs for skin cancer diagnosis

Frontier multimodal LLMs scored 42% on a 200-lesion skin cancer test where a small fine-tuned CNN scored 62%. Before routing medical imaging to a general LLM, benchmark it against a specialist model.

• multimodal — A model that works with more than text — reading images, audio, or video, and sometimes generating them too.
• LLM — A large language model — the neural network behind tools like Claude and ChatGPT, trained on huge amounts of text to predict what comes next.
• benchmark — A standard public test set for comparing AI models — the shared scoreboards behind every "model X beats model Y" claim.

• articleLlm Novice UpliftScale AI blog
• articleHealthcare Blind Spots: AI Models Prone To Fabricating DiagnosesMallory Lindahl
• articleInstitution-Specific LLM Prompting Recovers PHI That De-identification Systems and Their Gold Standards Both MissDaniel Palacios, Matthew Brady Neeley, Angel Adetomike Otto, Shalini Dhamodharan, John P. Woodhouse, Chi-fan Lin, Mark Zobeck, Zhandong Liu, Hyun-Hwan Jeong

Checking sign-in…

Loading comments…