Samsung Research has detailed two new artificial intelligence models designed to identify health patterns from data collected by wearable devices, as the technology company expands its research into AI-powered continuous health monitoring.
The research models, called xMAE and HiMAE, analyse biosignals such as heart activity, movement and sleep-related measurements gathered by devices including smartwatches.
Rather than examining individual measurements in isolation, the models are designed to help AI identify relationships and patterns across multiple biological signals, Samsung said.
The research forms part of Samsung’s broader push towards personalised digital health. In July, the company introduced a beta version of Samsung Health Assistant, which uses health data to generate personalised wellness recommendations.
Samsung has also outlined a wider “Connected Care” vision, aimed at combining AI with data continuously collected from wearable devices to provide more personalised health insights.
The two models are currently research projects rather than medical products.
xMAE is designed primarily to understand relationships between different heart-related signals, while HiMAE analyses patterns occurring across different time scales — ranging from rapid changes such as individual heartbeats to longer-term trends associated with sleep and physical activity.
Samsung said HiMAE can operate on a smartwatch-class processor in less than a millisecond, potentially allowing sophisticated analysis to take place directly on wearable devices.
The research contributes to the emerging field of health foundation models, in which AI systems learn from large volumes of biological data and can subsequently be adapted to perform a variety of health-related analytical tasks.