Using ChatBots and AI Agents in Clinics?

[drusso]

---

Members do not see this ad.

Anyone doing this for routine procedure follow-ups or med refills?

A.I. Chatbots Defeated Doctors at Diagnosing Illness

A small study found ChatGPT outdid human physicians when assessing medical case histories, even when those doctors were using a chatbot.

www.nytimes.com

 

[Ducttape]

50 participants that were paid. 26 were attendings, 24 residents.

not much difference between the 2 groups.

the article is basically focusing on the secondary analysis, where the LLM (large language module) was utilized to diagnose.

why? maybe because of:

This may be explained by the sensitivity of LLM output to prompt formulation.32

in the secondary analysis, my understanding is the LLM did better because doctors might have had preconceived notions on their diagnosis and instead of being entirely neutral to a a symptom/prompt, doctors focused on what would prove or disprove their notions.