Hacker News Reader

Who wants to do that? I assume you're talking about large language models? They are great at making word soup, but not great at diagnosis.

There have been expert systems since the 90's that perform better on average than doctors when diagnosing illness in patients, and / or help doctors make better decisions.

View on HN

>They are great at making word soup, but not great at diagnosis.

No they are.

>There have been expert systems since the 90's that perform better on average than doctors when diagnosing illness in patients,

No there has not been.

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10425828/

66% accuracy is not "great" and definitely not the best there's ever been.

If they aren't great then neither are doctors and the word becomes meaningless.

Context is important. 66% accuracy on cases that stumped doctors and took extensive testing is not trivial. It's not the same as 66% accuracy for everyday diagnosis.

The study references "clinicians" rather than "doctors." Clinicians include psychologists, pharmacists, nurses, physicians, paramedics, etc.

Point still stands.

Expert system diagnostic tools aren't that great.

Reply on news.ycombinator.com