Post by ARISE

A new study in Science from our group at Stanford ARISE received widespread attention and turned into an all too familiar headline: "Did AI beat ER doctors at diagnosis?" It's an eye-catching headline. And, it's also the least interesting thing about the study. Three quick corrections, courtesy of a Forbes piece on this article by an emergency medicine physician: 1) First, the doctors in the study weren’t ER doctors. They were internal medicine doctors, who have different training and focus. 2) The primary goal of emergency medicine is not always about landing on the precise diagnosis. It’s about ruling out life threats, managing uncertainty and moving patients safely through a high-volume, high-stakes environment. This differs from the sole goal of accurate diagnosis. 3) A text-based diagnostic exercise such as in this study, however well designed, doesn’t capture how real-life emergency medicine works. The AI was not practicing emergency medicine. It was offering a written opinion. Overall, the author-who is an emergency medicine physician-posits that the most important finding of the study in Science is not the comparative accuracy but rather it’s the fact that AI performed so well on messy, real-world, unprocessed clinical data.