Just took a look at this today and I agree that was interesting. Would be nice to have a resource to track different human agreement numbers and come up with some standards for comparisons to humans when reporting AI results.
What comes through in the @NatureMedicine article by @EricTopol is that one of the biggest contributions of #ArtificialIntelligence to medicine is highlighting how often physicians disagree with one another. A few examples from the article below. https://www.nature.com/articles/s41591-018-0300-7 …pic.twitter.com/jTTu0wvOJo
-
-
-
cc
@mandl KappaDB would be a big contribution to medical decision making
End of conversation
New conversation -
-
-
Eric highlights finding from our paper that most others missed. It’s staggering how little information is out there on ophthalmic diagnostic accuracy - likely much much more variable than we imagine! Upcoming work by
@sktywagner will explore in more detailThanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
This is exactly what we reported on the Health Informatics Journal a few days ago. Its implications are so far-reaching that we called this phenomenon, the "elephant in the record".https://journals.sagepub.com/eprint/3zKbhdu8CJCZrHfDbk3P/full …
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Every great advancement or innovation has always started with disagreement or disappointment.. eventually they all catch up.. AI is to stay
@EricTopolThanks. Twitter will use this to make your timeline better. UndoUndo
-
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.