Bien qu'il s'agisse d'une étude réalisée avec des comédiens en environnement contrôlé, l'intégration de la vidéo temps réel via Project Astra accélère les travaux de Google dans la téléprésence médicale automatisée.
Cette avancée marque le passage des assistants médicaux textuels passifs à des agents multimodaux capables de mener un examen physique virtuel direct et d'analyser le comportement non verbal en temps réel.
Basé sur Gemini et Project Astra avec une architecture multi-agents, le système AMIE analyse désormais les indices visuels et sonores comme la toux ou la démarche. Lors d'une étude randomisée sur des consultations simulées, des évaluateurs médicaux ont jugé ses performances diagnostiques et sa qualité de communication comparables à celles de médecins généralistes.
Impact concret
Les directeurs de l'innovation médicale et éditeurs de logiciels de télémédecine doivent intégrer l'arrivée d'agents vidéo multimodaux dans la conception de leurs futurs parcours de pré-diagnostic.
Applications concrètes
Analyse automatisée des symptômes visuels et auditifs à distance, guidage interactif des patients lors de l'auto-examen physique à domicile.
Breadcrumb Home Innovation & AI Models & research Google Research AMIE, our research medical AI system, demonstrates real-time clinical video consultation capabilities in a first-of-its-kind study. Anil Palepu Research Lead Aug 11, 2026 When you visit a doctor, a consultation extends far beyond words — a physician notices a cough, observes gait, or registers visible signs of discomfort. Today, Google Research and Google DeepMind are advancing AMIE, our research medical AI system, toward real-time clinical video consultations, with a first-of-its-kind demonstration of expert-level AI capabilities in this setting. Built on Gemini and Project Astra using a multi-agent architecture, AMIE now interprets visual and auditory cues, guides virtual physical exams, and reasons diagnostically in real time. In a randomized study using simulated consultations with patient actors and a group of primary care physicians, clinical evaluators assessed AMIE favorably across core clinical competencies, including history-taking thoroughness, diagnostic accuracy, management appropriateness and communication quality. Patient actors also preferred the video experience over text chat. AMIE remains a research system and more research is needed before responsible real-world clinical deployment, but it offers an exciting glimpse into the future of health AI. Read more on the Google Research blog . POSTED IN: Google Research Google DeepMind Google Health AI Related stories Google Health Google Health announces a strategic partnership with Abbott, a leader in health and wellness. By Rishi Chandra Google Ads Evolve your marketing with new AI tools By Josh Moser Gemini models See what 5 builders are making with Gemini Omni By Lindsey Lanquist Google DeepMind Our WeatherNext 2 AI model demonstrated a massive leap forward in predicting cyclones. Breadcrumb Home Innovation & AI Models & research Google Research AMIE, our research medical AI system, demonstrates real-time clinical video consultation capabilities in a first-of-its-kind study. Anil Palepu Research Lead Aug 11, 2026 When you visit a doctor, a consultation extends far beyond words — a physician notices a cough, observes gait, or registers visible signs of discomfort. Today, Google Research and Google DeepMind are advancing AMIE, our research medical AI system, toward real-time clinical video consultations, with a first-of-its-kind demonstration of expert-level AI capabilities in this setting. Built on Gemini and Project Astra using a multi-agent architecture, AMIE now interprets visual and auditory cues, guides virtual physical exams, and reasons diagnostically in real time. In a randomized study using simulated consultations with patient actors and a group of primary care physicians, clinical evaluators assessed AMIE favorably across core clinical competencies, including history-taking thoroughness, diagnostic accuracy, management appropriateness and communication quality. Patient actors also preferred the video experience over text chat. AMIE remains a research system and more research is needed before responsible real-world clinical deployment, but it offers an exciting glimpse into the future of health AI. Read more on the Google Research blog .