Medical AI ushered in a major breakthrough: the research system AMIE demonstrated real-time clinical video consultation capabilities for the first time. In real medical scenarios, doctors’ diagnosis often not only relies on the patient’s verbal description, but also keenly captures signs of discomfort such as coughing, abnormal gait, or facial discomfort.

Medical AI ushered in a major breakthrough: the research system AMIE demonstrated real-time clinical video consultation capabilities for the first time. In real medical scenarios, doctors’ diagnosis often not only relies on the patient’s verbal description, but also keenly captures signs of discomfort such as coughing, gait abnormalities, or facial discomfort. Recently, in the field of medical artificial intelligence, the research-based medical AI system AMIE jointly promoted by Google Research and Google DeepMind has achieved a new technological breakthrough, demonstrating expert-level application capabilities in real-time clinical video consultation scenarios for the first time. The system is based on the underlying architecture of Gemini and Project Astra, and adopts a multi-agent collaboration model. In the latest functional upgrade, AMIE can not only accurately interpret and combine visual and auditory clues, but also flexibly guide virtual physical examinations and carry out real-time diagnostic reasoning, making AI closer to the consultation logic of human experts in remote medical interactions. In a recent randomized controlled study of simulated consultations, researchers arranged for volunteers playing patients to interact with primary care physicians and the AI ​​system. The clinical evaluation results show that professional evaluators gave positive recognition to AMIE in a number of core clinical capabilities, including the comprehensiveness of medical history collection, accuracy of diagnosis, appropriateness of treatment plans, and overall communication quality. At the same time, patient actors generally reported after the experience that compared to traditional text chat interaction, the video consultation format is more in line with clinical communication habits and the experience is more friendly. The R&D team pointed out that AMIE is still in the academic research stage and still needs to undergo a large amount of rigorous verification and in-depth research before it can truly move towards responsible real-world clinical deployment. Nonetheless, this progress still outlines a promising technological picture for future intelligent development in the medical and health field. via AI News (author: AI Base)