Biosignals Models

Biosignal foundation models learn from continuous physiological and wearable sensor streams — continuous glucose monitoring (CGM), electrocardiography (ECG), electroencephalography (EEG), photoplethysmography (PPG), actigraphy, and related time-series. By pretraining on long, noisy, multi-channel recordings, they capture the temporal patterns that underlie cardiac, neural, metabolic, and sleep dynamics, enabling downstream tasks like arrhythmia detection, seizure prediction, and glucose forecasting. As wearables make these signals ubiquitous, such models are extending foundation-model methods from molecules and images to the continuous data of everyday physiology.

1 model in this category