Google Research pairs wearable health age with simulated habit coaching

Google Research published two papers describing an investigational machine learning framework that turns passive wearable data into a "coachable health age" and simulates personalized 12-week habit plans. The health-age model draws on resting heart rate, daily steps, sleep duration and weekly Active Zone Minutes, with optional markers such as VO2max and insulin resistance, and was tested across the All of Us program, a longitudinal Fitbit cohort and the WEAR-ME study.