7 months ago
AI Model Predicts 130 Diseases from One Night's Sleep
Scientists at Stanford Medicine created an AI called SleepFM that can predict over 130 diseases by analyzing just one night of sleep.
They trained it using sleep data from 65,000 people, including brain activity, heart rate, and breathing.
The AI is good at predicting serious diseases like Parkinson’s, cancer, and heart problems.
The researchers want to make it even better by adding data from wearable devices.
This could help doctors find health issues early, but they are still figuring out exactly how the AI makes its predictions.
Stanford Medicine developed SleepFM, an AI model predicting 130 diseases from one night of sleep data.
Trained on 600,000 hours of sleep data from 65,000 participants, including brain and heart activity.
Achieved high accuracy (C-index > 0.8) in predicting diseases like Parkinson’s, cancer, and heart conditions.
Researchers plan to improve the model by incorporating data from wearable devices.
SleepFM is one of the first AI models to use sleep data for disease prediction.
- Who
- Researchers from Stanford Medicine
- What
- Developed an AI model called SleepFM that predicts disease risk from sleep data
- Where
- Stanford Medicine, United States
- When
- Data collected between 1999 and 2024, study published in Nature Medicine
- Why
- To provide early disease prediction using sleep data
Key facts
- AI Model Name
- SleepFM
- Developed By
- Stanford Medicine
- Training Data
- 600,000 hours of sleep data from 65,000 participants
- Diseases Predicted
- 130 diseases
- Accuracy Metric
- C-index > 0.8 for many diseases
- Key Predictions
- Parkinson’s, dementia, heart attack, cancers, and more
- Future Improvements
- Adding wearable data to enhance accuracy
Quotes
Dr James Zou
Associate professor of biomedical data science and co-author of the study
“From an AI perspective, sleep is relatively understudied. There’s a lot of other AI work that’s looking at pathology or cardiology, but relatively little looking at sleep, despite sleep being such an important part of life.”
indianexpress.com
“It doesn’t explain that to us in English. But we have developed different interpretation techniques to figure out what the model is looking at when it’s making a specific disease prediction.”
indianexpress.com
Dr Emmanuel Mignot
Co-author of the study
“We record an amazing number of signals when we study sleep. It’s a kind of general physiology that we study for eight hours in a subject who’s completely captive. It’s very data rich.”
indianexpress.com
Sources
Related news
Shingrix Linked to Lower Heart Attack and Stroke Risk
Dairy Gains Attention in India’s Growing Fitness Movement
AI Enters IVF Labs as Decision-Support, Not Embryologist Replacement
Study says Congo-Uganda Ebola outbreak began with new animal transmission
