desk
Share
News Topical, Digital Desk : Artificial Intelligence (AI) is rapidly gaining popularity. There's hardly any field where AI isn't being used. AI is also being used in the medical field for numerous research and inventions. Recently, scientists have taken a revolutionary step in this direction.
In fact, scientists have created an AI model that can predict the risk of 130 diseases and death by analyzing just one night's sleep data. Yes, a night's sleep will not only alleviate fatigue but also become a window into your future, thanks to AI.
What is the 'Sleep FM' model?
According to recent research published in the journal Nature Medicine, this AI model has been named "SleepFM." It was developed by scientists from Stanford University in the US and several other prestigious institutions. The model's most significant strength is its accuracy. During testing, its C-index was found to be over 0.75, meaning that more than 75% of its predictions were correct.
How does this system work?
This AI model is based on polysomnography (PSG) data. PSG is a detailed sleep study trained on over 585,000 hours of sleep records from 65,000 people collected from Stanford's sleep clinic.
According to Professor Emmanuel Mignot, senior author of the research, a person is in complete control during sleep, and their body gives off numerous signals. The AI model understands the hidden physiology of sleep by collecting four main types of data:
- Brain activity – through EEG and EOG
- Heart rate – via ECG or EKG
- Muscle activity – through EMG
- Respiratory signals – based on breathing patterns and changing signals over time.
Identification and accuracy of major diseases
Researchers studied health records for over 1,000 categories of illnesses, of which the AI identified 130 illnesses with high accuracy, with a score of 0.81 meaning the AI correctly identified them eight out of 10 times.
Read More: Your New Year's weight loss resolution won't be broken! Motivate yourself with these 5 tips
--Advertisement--
