@jonc101x @StanfordCompMed Dr. Chen, DECIDE-AI separates early live evaluation from definitive efficacy trials. What evidence should trigger re-evaluation after a model update: changed outputs, changed clinician behavior, or both? https://t.co/lH7sSR9oX2
So far, Altmetric has seen
157 X posts from 141 X users, with an upper bound of 3,146,578 followers.
15 followers
122 followers
@ScienceMagazine RADAR performed strongly across retrospective datasets, but benchmark accuracy is not yet evidence of improved care. So “expert-level” describes tested tasks, not deployment. Which live clinical outcome should be tested first? Sources: …
1,937 followers
https://t.co/WwtmG2LbJY 人間の判断機会は、「残さなければいけない」し、AIと使用者との関係がどのように想定されるかを、開発時から意識する必要があります。 DECIDE-AIガイドラインにも、その辺の記述がかなりしっかりあります。 https://t.co/9zY3iX5J8z
25,792 followers
RT @ApBE_: Guía de informes para la evaluación clínica en etapa temprana de los sistemas de apoyo a la toma de decisiones impulsados por…