James Zou

Stories

AI cracks the sleep code: one night predicts 130 diseases

Co-senior author of SleepFM study

Stanford researchers trained an AI on 600,000 hours of people sleeping. SleepFM analyzes brain waves, heartbeats, and breathing from a single night and predicts your risk for 130 diseases—dementia, heart attacks, cancer, mental disorders—with over 80% accuracy. The breakthrough turns a night of sleep into a comprehensive disease-screening tool.

Updated May 19

Notable Quotes

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.

from AI cracks the sleep code: one night predicts 130 diseases · Updated May 19

SleepFM is essentially learning the language of sleep.

from AI cracks the sleep code: one night predicts 130 diseases · Updated May 19