In plain words: Instead of learning from raw sensor readings, it learns from behavioral signals like sleep and activity, drawn from wearables on hundreds of thousands of people. Across 57 health tasks it beat raw-sensor models, especially on sleep prediction, and did best when both were combined.
Abstract · Beyond Sensor Data: Foundation Models of Behavioral Data from Wearables Improve Health Predictions
Wearable devices record physiological and behavioral signals that can improve health predictions. While foundation models are increasingly used for such predictions, they have been primarily applied to low-level sensor data, despite behavioral data often being more informative due to their alignment with physiologically relevant timescales and quantities. We develop foundation models of such behavioral signals using over 2.5B hours of wearable data from 162K individuals, systematically optimizing architectures and tokenization strategies for this unique dataset. Evaluated on 57 health-related tasks, our model shows strong performance across diverse real-world applications including individual-level classification and time-varying health state prediction. The model excels in behavior-driven tasks like sleep prediction, and improves further when combined with representations of raw sensor data. These results underscore the importance of tailoring foundation model design to wearables and demonstrate the potential to enable new health applications.
Eray Erturk, Fahad Kamran, Salar Abbaspourazad, Sean Jewell, Harsh Sharma, Yujie Li, Sinead Williamson, Nicholas J Foti, Joseph Futoma
arXiv:2507.00191 · cs.LG, cs.AI · submitted Jun 30, 2025
abstract · pdf · html · Accepted to ICML 2025
They find high accuracy in detecting many conditions: diabetes (83%), heart failure (90%), sleep apnea (85%), etc.