Sleep trackers and AI used to predict early signs of Alzheimer’s

The grant was awarded to Joyita Dutta, a professor of biomedical engineering at the University of Massachusetts Amherst, for a five-year study to assess whether wearable sleep trackers can predict blood biomarkers of Alzheimer’s disease.

Sleep disruption is one of the hallmarks of Alzheimer’s disease, even before cognitive symptoms manifest. However, gold-standard sleep assessments are expensive and usually only provide data from a single night.

With the goal of expanding early Alzheimer’s diagnosis, this study will assess if unobtrusive sleep trackers can log sleep patterns that correlate with future cognitive decline, as indicated by blood biomarkers.

“Many people already wear smartwatches to sleep these days. Imagine receiving an alert from your smartwatch advising you to see a neurologist. That could be the direction we are headed,” Dutta said in a statement.

Her study will evaluate the sleep patterns of people with a genetic predisposition to developing Alzheimer’s disease, but no observable signs of cognitive impairment.

Instead of completing a one-night sleep study, the participants will also wear three types of sleep trackers for a week: an Apple Watch, the Oura Ring and CGX Patch, a wearable electroencephalogram (EEG) that is essentially a sticky forehead patch with metal electrodes that can measure brain activity.

The data from these wearables will be compared to new blood tests measuring amyloid and tau proteins, key early biomarkers of Alzheimer’s disease. This assessment will be repeated after two years to detect possible changes.

“Our previous work includes developing AI-based predictive models connecting sleep patterns to cognitive impairment. This grant allows us to take that research to the next level,” added Dutta.

“The project will enable the integration of a wealth of new data — genetic information, wearables-derived metrics, and blood-based biomarkers to create a more comprehensive picture of the sleep-dementia axis.”