A simple overnight sleep test combined with artificial intelligence may one day identify people at higher risk of developing dementia years before memory problems begin.
Researchers developed a machine learning-based tool that estimates a person's "brain age" using sleep electroencephalography (EEG), a test that records electrical activity in the brain during sleep. They found that participants whose brains appeared older than their actual age were significantly more likely to develop dementia during follow-up, suggesting the technology could eventually become an early digital marker for cognitive decline.
The findings come from an individual-participant data meta-analysis involving 7,105 adults from five long-running community studies across the United States, making it one of the largest investigations to date examining whether sleep-related brain activity can forecast dementia risk.
AI-linked accelerated brain aging with future dementia
The researchers tracked participants who did not have dementia at the time of their overnight sleep studies. Using AI, they calculated each person's Brain Age Index (BAI), which measures the difference between brain age estimated from sleep EEG patterns and chronological age.
Across all five cohorts, every 10-year increase in BAI was associated with a 39% higher risk of developing dementia, even after accounting for age, sex, education, lifestyle factors and overall health.
The association remained significant after researchers adjusted for additional factors known to influence dementia risk, including depression, diabetes, hypertension, stroke, sleep apnea severity, baseline cognitive performance and theAPOE ε4 genetic variant linked to Alzheimer's disease.
According to the study, "each 10-year increase in BAI was associated with a 39% higher risk of incident dementia," independent of many established dementia risk factors.
Sleep signals may reveal changes before memory problems emerge
Unlike traditional sleep measures that focus on how long people spend in different sleep stages, the AI analyzed subtle EEG "microstructures" reflecting brain activity during sleep, including sleep spindles, slow-wave activity, and other electrical patterns associated with healthy brain aging.
The researchers said these microscopic changes may capture neurophysiological alterations long before noticeable cognitive symptoms appear.
"Our findings suggest that sleep EEG-based BAI may serve as a promising early digital marker for dementia risk stratification," the authors wrote.
Most participants were cognitively normal when their sleep studies were performed, yet higher brain age still predicted future dementia over follow-up periods ranging from about four to nearly 17 years, depending on the study.
Promising but not ready for routine care
While the results highlight the growing role of AI in preventive neurology, the researchers cautioned that the technology is not ready for widespread clinical use.
The analysis shows an association, not that accelerated brain aging detected during sleep causes dementia. The authors also noted that dementia diagnoses differed somewhat across the five cohorts, and additional studies are needed to validate the tool in broader populations.
Future research will also explore whether wearable EEG devices could eventually estimate brain age outside specialized sleep laboratories, making the technology more accessible.
The researchers emphasized that the Brain Age Index should be viewed as a risk assessment tool rather than a diagnosis.
"Elevated sleep EEG-based BAI, a machine learning marker of brain aging, was independently associated with higher dementia risk in community-based populations," the authors concluded. "With future validation in wearable EEG devices, BAI has the potential to complement emerging plasma and imaging markers, supporting multimodal risk stratification to better inform clinical decision-making."
For now, the findings underscore the growing importance of sleep as a window into brain health, and suggest that AI could eventually help identify people who may benefit from closer monitoring or earlier interventions long before dementia symptoms become apparent.
