Researchers have proposed the Personal Health Agent (PHA), a research framework designed to reason about multimodal data and provide personalized, evidence-based guidance. The work looks at how large language models (LLMs), combined with data from wearable devices, could support personal health journeys while handling different types of health questions.
The framework uses a multi-agent architecture that breaks personal health and wellness support into three core roles: data science, domain expert, and health coach. Each role is handled by a specialist sub-agent.
To evaluate the system, the researchers used a real-world dataset from an IRB-reviewed study involving ~1200 users who gave informed consent to share data from Fitbit, a health questionnaire, and blood test results. They ran automated and human evaluations across 10 benchmark tasks.
The evaluation included more than 7,000 annotations and 1,100 hours of effort from health experts and end-users. The researchers said the work represents the most comprehensive evaluation of a health agent to date.
The source says the framework is for research purposes and should not be considered a description of any specific product, service, or feature currently in development or available to the public.
Source: research.google.
Companies can share verified announcements through Newz9’s international press release submission page.
