Can an AI Brain-Age Score Predict Dementia

How Can AI Estimate Brain Age From a Sleeping EEG?

Electroencephalography, or EEG, records electrical activity through electrodes placed on the scalp. During sleep, the brain produces recognizable rhythms and transitions between sleep stages. These signals change with normal aging, giving researchers an opportunity to train algorithms that estimate a person's neurological age from features distributed across an overnight recording.

The difference between estimated brain age and chronological age is often called a brain-age gap. A brain that appears older on an algorithm is not necessarily diseased: medications, sleep apnea, disrupted sleep, neurological conditions and technical differences between recording systems may all influence EEG patterns. The score therefore needs clinical context and should not be interpreted as a biological expiration date.

Could an Older Brain-Age Score Reveal Early Dementia Risk?

Dementia is a syndrome caused by several diseases, including Alzheimer's disease, vascular brain injury and Lewy body disease. Changes can begin years before daily functioning is noticeably impaired, which is why researchers are investigating blood biomarkers, imaging, cognitive testing and digital signals alongside EEG. A sleep-based measure could be attractive because many people already undergo EEG monitoring during clinical sleep studies.

Association is not the same as prediction. Researchers must demonstrate that a brain-age score adds useful information beyond age, health history, sleep disorders and established cognitive assessments. Prospective studies are also needed to determine whether people with an older-appearing EEG are actually more likely to develop cognitive impairment over time and whether acting on the result improves health outcomes.

What Must Happen Before EEG Brain-Age Testing Reaches Clinics?

An algorithm developed in one dataset may perform differently in another hospital, age group or population. External validation should examine calibration, false-positive results and performance across demographic and clinical groups. Researchers must also determine how sleep deprivation, sedating medicines, epilepsy, stroke and sleep-disordered breathing affect the estimated age.

Even a technically accurate score needs a defined clinical purpose. It might eventually help clinicians decide who should receive formal cognitive testing or specialist assessment, but it should not replace a medical history, neurological examination or validated evaluation of memory and daily functioning. Until clinical utility is established, consumers should be cautious about commercial tests claiming to calculate brain age or forecast dementia from limited physiological data.