Poor sleep can cause heart disease, cancer or dementia; this new AI model helps spot the symptoms early

Not getting a good sleep through the night can negatively impact your health. (Picture Credit: Pexels)

• Stanford AI predicts risk of 130 diseases using detailed sleep data
• SleepFM AI analyzes brain, heart, and breathing signals to detect diseases early.
• Future versions may use wearables for at-home health monitoring

Did our AI summary help?

Not getting a good sleep through the night can negatively impact your health. According to researchers at Stanford Medicine, a single night’s sleep could quietly reveal whether your body is heading towards serious illness long before doctors ever spot it.

In a recent study scientists have trained an artificial intelligence system to predict the risk of more than 100 diseases using detailed sleep data. From cancer and dementia to heart attacks and mental health disorders, the model hunts for tiny mismatches between the brain, heart and breathing that the human eye usually misses. The research suggests sleep is less of a passive shutdown and more of a nightly health check we’ve been ignoring.

The AI, called SleepFM, was fed nearly 600,000 hours of overnight sleep recordings taken from 65,000 people across decades. These weren’t basic fitness-tracker stats, but full polysomnography tests, the kind involving wires, sensors and a lab bed, tracking everything from brain activity and heart rhythms to eye flickers and airflow. Traditionally, doctors use only a fraction of this data to diagnose sleep disorders. The rest? Filed away, unexplored, until now.

Also read: Doctor says these 3 medicines can save a heart attack patient

Stanford sleep specialists realised they were sitting on a physiological goldmine. For eight hours straight, the body reveals how its systems sync up, or don’t. With AI finally capable of handling the sheer complexity, researchers say they’ve uncovered patterns that hint at future disease years before diagnosis.

Much like language models learn words and sentences, SleepFM learned sleep in five-second pieces. It was trained to understand how the brain, heart and lungs talk to each other overnight. When one signal goes missing, the AI learns to fill in the gaps, a trick that helps it spot when the body’s systems fall out of rhythm.

When researchers linked sleep data to long-term medical records, the results showed that AI could reliably flag risks for 130 conditions. It was especially strong at predicting Parkinson’s disease, dementia, heart attacks, breast and prostate cancer, and even overall mortality, often with accuracy already considered useful in clinical medicine.

The clearest danger signs appeared when body systems were out of sync, a brain deeply asleep while the heart behaved as if it were wide awake, for instance. Researchers believe future versions could work with wearable devices, turning ordinary nights at home into powerful early-warning systems.

Also read: Third-hand smoking puts children and elderly at home at risk, expert lists tips to protect against its toxic effect

Disclaimer: This article, including health and fitness advice, only provides generic information. Don’t treat it as a substitute for qualified medical opinion. Always consult a specialist for specific health diagnosis.