Khaberni - A recent study indicated the possibility of using sleep data to predict the risk of diseases, thanks to new developments in the field of artificial intelligence.
Researchers from the Stanford University School of Medicine developed an artificial intelligence model that was trained on approximately 600,000 hours of sleep data collected from more than 60,000 participants in various sleep clinics.
According to "Fox News," the SleepFM program appears to be capable of predicting a person's risk of more than 100 health conditions, including dementia and heart disease, as reported by the researchers.
"Sleep Me"
The researchers trained the "Sleep Me" model using polysomnography, a comprehensive sleep measurement that tracks brain and heart activity, as well as breathing, leg and eye movements. They noted that this measurement is considered the gold standard for sleep studies.
Dr. James Zou, the study’s co-lead author, said: "Sleep contains much more information about future health than we currently use."
He added: "By learning the language of sleep, our artificial intelligence model opens new horizons for studying the science and medicine of sleep," pointing out that humans spend about a third of their lives sleeping.
In this study, the team linked sleeping data with the electronic health records of the participants, which provided data up to 25 years.
Predicting 130 Diseases
Through the analysis of data and disease categories in those health records, the artificial intelligence model discovered 130 diseases it could predict with reasonable accuracy.
Zou said: "By analyzing one night of sleep using advanced artificial intelligence, we found that sleep patterns can predict the risk of more than 100 different diseases years before diagnosis."
These diseases included dementia, heart diseases, stroke, kidney diseases, and even mortality rates.
The researchers noted that the predictions of the model were particularly strong in relation to cancers, pregnancy complications, circulatory system diseases, and mental disorders.
The researchers hope to expand the scope of research to include collecting data from patients using wearable devices, which may help in determining what the model interprets accurately.
