MotionAge: A Deep Learning Framework for Biological Age Prediction from Wearable Activity

Population aging has created a growing need for scalable measures of health beyond chronological age. Existing approaches to biological age often rely on clinical or laboratory data, limiting their ability to capture real-world, continuously evolving health status. We propose MotionAge, a deep learning framework that learns a mortality-calibrated biological age directly from high-frequency wearable activity data. The approach combines deep sequence models with a wear-aware modeling strategy that explicitly encodes observation reliability, enabling robust representation learning from noisy and irregularly observed time series. In NHANES accelerometer data, MotionAge improves mortality discrimination over chronological age and established benchmarks. These results highlight the potential of wearable data to enable scalable, real-time assessment of aging in population settings.