Health prediction from sleep data
Earlier, a disease was detected only when the condition became serious. But due to advances in science, it is now possible for us to know which diseases can affect us, what their severity is, and what diseases may occur in the future. Researchers have developed an artificial intelligence model, called SleepFM, that can predict a person’s risk for 130 potential diseases based on sleep data. The model has been created by researchers from various institutions including Stanford University in the US and has been trained on approximately 600,000 hours of sleep data collected from 65,000 individuals. The results of this study have been published in the medical journal Nature Medicine. Let us know how this device works and how it can predict future diseases.
Working of SleepFM
This AI system was initially tested to identify common sleep-related problems, such as tracking different stages of sleep or assessing the severity of sleep apnea. After this, the sleep data was combined with the patients’ health records to see which diseases they might be at risk for in the future. The researchers reported that the model was able to accurately predict 130 diseases out of more than 1,000 diseases present in health records.
health signs in sleep
According to Emmanuel Mignot, professor of sleep medicine at Stanford University, “During sleep, many signals from the body are recorded. The body’s normal activities are studied for up to eight hours, making the data very rich.”
data collection process
Polysomnography was used to analyze sleep, which is considered the most reliable method for studying sleep. In this, many signals are recorded using sensors, such as:
brain activity
heartbeat
breathing pattern
eye movement
Muscle activity. SleepFM analyzes all these data streams together and understands their interrelationships.
New method of AI training
The team used a ‘leave-one-out’ contrastive learning technique to train the AI. In this method, one type of data is deliberately hidden, and the AI is challenged to guess the missing information based on the remaining signals. This improves model understanding and accuracy.
identification of diseases
The research found that this AI is particularly adept at predicting a variety of diseases, including:
cancer
Pregnancy related problems
Diseases related to heart and blood flow
Mental health problems. In many cases, its C-index score was greater than 0.8, indicating good predictive accuracy. According to the researchers, the diseases whose risk SleepFM can predict from just one night’s sleep data include:
dementia
heart attack
heart failure
chronic kidney disease
stroke
Atrial fibrillation. Additionally, the model has also proven effective in predicting the risk of diseases like Parkinson’s and growth-related problems in children. Overall, this research shows that your sleep is not just a means to relieve fatigue, but it is also a mirror of your future health.
Comments are closed.