Researchers from the Digital Health Team at Samsung Research America are developing new AI technologies to continuously understand a person's health state from biosignals, generate health insights, and offer appropriate health guidance.[reference:24] Samsung researchers recently introduced two foundation models based on wearable data: xMAE and HiMAE.[reference:25]
xMAE: Learning Cardiac Signals from PPG
xMAE (Physiology-Aware Masked Cross-Modal Reconstruction for Biosignal Representation Learning) learns the temporal relationship between different biosignals.[reference:26] An electrocardiogram (ECG) on wearables directly measures the electrical activity of the heart, but typically requires users to pause and take an active measurement. Photoplethysmography (PPG) can indirectly measure cardiac function by detecting changes in blood flow and can be passively and continuously measured.[reference:27]
xMAE is a biosignal pretraining framework designed to learn the temporal relationship between the two signals by reconstructing masked portions of the ECG signal using the more easily and continuously measured PPG signal. As a result, cardiovascular health-related features can be analyzed more precisely using PPG without requiring separate manual ECG measurements.[reference:28]
HiMAE: Understanding Health Patterns Across Time Scales
HiMAE (Hierarchical Masked Autoencoder) understands health patterns across different time scales in wearable time-series data.[reference:29] Samsung Research has developed HiMAE to be small enough to run entirely on smartwatch-class hardware, achieving inference in less than a millisecond on smartwatch-class processors.[reference:30]
Both studies demonstrate advancement of health AI models that can better understand the physiological relationships and temporal structure of biosignals.[reference:31] Samsung's work on xMAE and HiMAE was accepted to the International Conference on Machine Learning (ICML) and the International Conference on Learning Representations (ICLR), respectively.[reference:32]
The Future of Preventive Health
These health foundation models use self-supervised learning to learn meaningful features from unlabelled biosignal data. After being pretrained on large-scale health data, the models can be applied to perform a wide range of downstream health tasks, including biosignal analysis, developing new or improved biomarkers, and health issue prediction.[reference:33] This technology is helping power Samsung's Connected Care vision — a future where care moves from reactive treatment toward preventive, personalized, and connected experiences.[reference:34]