AI-Powered Sleep Data Revolutionizes Health Predictions

Researchers have developed an AI model, SleepFM, that uses sleep data from polysomnography to predict the risk of over 100 health conditions. The model, tested on data from 65,000 participants, shows promise in accurately predicting diseases like dementia, heart conditions, and even certain cancers with over 80% accuracy.

In a groundbreaking study, researchers have unveiled an artificial intelligence model, SleepFM, capable of predicting more than 100 health conditions based on a single night's sleep lab data. Developed with insights from over half a million hours of polysomnography data from 65,000 individuals, the model is redefining health risk assessments.

According to the team, polysomnography serves as an underutilized yet invaluable repository of physiological data. Dr. Emmanual Mignot from Stanford Medicine highlighted that integrating AI with in-depth sleep studies could transform our ability to foresee critical health issues, effectively using sleep patterns as a predictive tool.

The model has already been tested on thousands of patients' records collected from the Stanford Sleep Medicine Center. It's shown significant accuracy, particularly in predicting life-altering health conditions such as dementia and heart diseases. Efforts are underway to enhance predictions by incorporating data from wearable technologies.

(With inputs from agencies.)