A new AI model, SleepFM, developed by Stanford University, predicts over 100 health conditions based on sleep data. Trained on 580,000 hours of data from 65,000 patients, it accurately predicts diseases like Parkinson's and Alzheimer's with over 80% accuracy. The model's predictions are based on various health signals recorded during sleep. Future improvements will include data from wearables, but the current study's sample may not represent the general population. Teljes cikk (Euronews.com)