SleepFM: Multi-modal Representation Learning for Sleep Across Brain Activity, ECG and Respiratory Signals

Summary

SleepFM (Stanford, James Zou Lab) is a multimodal sleep foundation model trained on 585,000+ hours of polysomnography (PSG) from 65,000+ participants — the largest sleep FM by training data volume at publication. Its dual publication at ICML 2024 (method) and Nature Medicine 2025 (clinical validation) represents the highest-impact wearable-adjacent FM publication in the field.