Researchers say AI can accurately predict 130 diseases from one night of sleep
Researchers say AI can accurately predict 130 diseases from one night of sleep
Publish Date: 2026-01-06 16:46:00
Source Domain: www.scrippsnews.com
- Researchers at Stanford University have developed an AI system called SleepFM which forecasts a person’s risk of developing certain diseases and even death by analyzing sleep data from just one night.
- The researchers analyzed sleep recordings from over 65,000 individuals to introduce their new AI tool.
- The SleepFM AI can predict risks for 130 diseases with more than 80% accuracy years before these conditions are diagnosed.
- Data used in the study includes over 585,000 hours of sleep recordings from participants aged 2 to 96.
- The AI was able to identify potential risks for a wide range of health conditions including cancer, cardiovascular issues, mental disorders, Parkinson’s disease, and dementia.
- Emmanuel Mignot, co-author of the study, emphasized that sleep data is rich in physiological information.
- James Zou, another co-author of the study, noted that sleep as a field is relatively understudied compared to other areas in AI, despite its significant importance.
- The findings underscore the importance of studying sleep in the context of advanced AI applications for health risk prediction.
