# Stanford AI Predicts 130 Diseases From a Single Night of Sleep

*2026-01-24 03:32 UTC · Science · 1 min read*

> Stanford's SleepFM AI analyzes one night of sleep data to predict 130+ diseases with up to 89% accuracy.

Stanford Medicine researchers have developed SleepFM, an AI model that predicts more than 130 health conditions from one night of sleep data. The system was trained on 585,000 hours of polysomnography recordings from 65,000 patients collected over 25 years.

The model achieved exceptional accuracy for Parkinson's disease (89%), prostate cancer (89%), breast cancer (87%), dementia (85%), and heart attack (81%). For all-cause mortality, SleepFM correctly ranked patient risk 84% of the time.

The key finding: misalignment between different body systems during sleep—such as a brain that appears asleep while the heart remains alert—signals higher disease risk. The research was published in Nature Medicine.

## Sources

- [news.stanford.edu](https://news.stanford.edu/stories/2026/01/ai-model-sleep-disease-risk-research-sleepfm)
- [med.stanford.edu](https://med.stanford.edu/news/all-news/2026/01/ai-sleep-disease.html)

---
TechTools News · https://techtools.cz/tools/news/article/118-stanford-ai-predicts-130-diseases-from-a-single-night-of-sleep
