Post by Google DeepMind

Our new research initiative AI co-clinician explores how multimodal agents might assist in new ways of delivering care. 🩺 It’s designed to function as a collaborative member of the care team that interacts with patients under expert supervision. We evaluated AI co-clinician in both clinical and patient-facing settings. In head-to-head blind testing on medical evidence generation, physicians consistently preferred its responses to leading evidence synthesis tools. Beyond this, we’re also investigating how AI co-clinician performs within patient-facing research contexts. It can process visual, auditory and sensory cues under clinician supervision to assist people in their care journeys. This means it could help analyze a patient's walk from video, listen to their breathing, or look at how a rash is appearing. Alongside physicians from Harvard Medical School and Stanford University School of Medicine, we created a study with 20 scenarios and patient-actors to assess Al co-clinician. In our evaluations, it demonstrated new capabilities beyond text-only systems - such as guiding patients through complex physical exams in real time. We’re further advancing this research with academics and institutions globally and will gradually expand our clinician-facing trusted tester program to additional sites - to understand more perspectives of health workers and patients worldwide. Find out more → https://goo.gle/4mXzZQY

Video Content