A new Nature Medicine paper introduces SleepFM, a foundation model trained on:

  • 585,000+ hours of sleep studies
  • 65,000 participants
  • EEG, ECG, respiratory + muscle signals From a single night of sleep, it predicted: 🧠 Dementia — C-Index 0.85 ❤️ Heart failure — 0.80 🫀 MI — 0.81 🩸 CKD — 0.79 ⚰️ All-cause mortality — 0.84 130 diseases achieved strong predictive performance (C-Index ≥ 0.75) https://lnkd.in/g97_yjsu . This isn’t just sleep staging. It’s a sleep foundation model that learns the physiological “language” of the brain, heart, and lungs — and translates it into future disease risk. Sleep may be the most underutilized biomarker in medicine. As AI + wearables scale, one night of sleep could become a powerful early-warning system for longevity. Now imagine integrating genomics, longitudinal multi-omics, and advanced imaging with AI — transforming medicine from reactive treatment to proactive prevention of age-related chronic disease. This is exactly what Human Longevity has been doing for over 10 years.
This article is for educational purposes only and is not a substitute for personalized medical advice, diagnosis, or treatment.