A1142
Title: Large-scale training of foundation models for wearable biosignals
Authors: Salar Abbaspourazad - Apple (United States) [presenting]
Abstract: Wearable devices can conveniently record various biosignals, creating the opportunity to monitor health status without disruption to one's daily routine. Despite the widespread use of wearable devices and existing digital biomarkers, the absence of curated data with annotated medical labels hinders the development of new biomarkers to measure common health conditions. To address this challenge, self-supervised learning has been employed using the unlabeled sensor data collected under informed consent from the large longitudinal Apple heart and movement study (AHMS) to train foundation models for two common biosignals: photoplethysmography (PPG) and electrocardiogram (ECG) recorded on Apple Watch. PPG and ECG datasets are curated from AHMS, that include data from ~141K participants spanning ~3 years. It is shown that the pre-trained foundation models readily encode information regarding participants' demographics and health conditions. To the best of knowledge, this is the first study that builds foundation models using large-scale PPG and ECG data collected via wearable consumer devices. Prior studies have commonly used smaller-size datasets collected in clinical and experimental settings. It is believed that PPG and ECG foundation models can enhance future wearable devices by reducing the reliance on labeled data and hold the potential to help users improve their health.
