AI Predicts Accelerated Brain Aging

Researchers at UC San Francisco and Beth Israel Deaconess Medical Center in Boston have developed a machine-learning model that analyzes brain activity during sleep to estimate a person’s “brain age” and identify elevated risks of developing dementia. Published in JAMA Network Open, the study analyzed EEG recordings from approximately 7,000 participants and found that a brain appearing older than a person’s actual chronological age increased dementia risk by nearly 40 percent for every 10-year discrepancy.

Machine-Learning Model Evaluates Microscopic Sleep Signals

The research team built their machine-learning model by combining 13 microscopic features extracted from electroencephalography, or EEG, brain wave recordings. They applied this system to information drawn from five separate studies involving participants ranging in age from 40 to 94. None of the individuals had dementia when their respective studies began. Researchers tracked the cohort for periods ranging from 3.5 to 17 years, during which about 1,000 participants developed dementia.

According to the findings, small and highly detailed patterns in sleeping brain waves provide information that standard sleep measurements fail to detect. Previous pooled analyses involving multiple participant groups found no meaningful association between dementia risk and common sleep metrics, such as total time spent in various sleep stages or sleep efficiency. “Broad sleep metrics don’t fully capture the complex multidimensional nature of sleep physiology,” said senior author Yue Leng, MBBS, PhD, an associate professor of psychiatry at the UCSF School of Medicine.

Brain Wave Patterns Connected to Memory and Risk

Several EEG patterns utilized in the brain-age calculation are already recognized for supporting memory and cognitive health. These include delta waves and sleep spindles. One of the study’s most notable findings involved large, sudden spikes in EEG signals known as kurtosis, which researchers associated with a lower risk of developing dementia.

The correlation between an older estimated brain age and a higher dementia risk remained statistically significant even after researchers adjusted for education, smoking, body mass index, physical activity, underlying medical conditions, and genetic risk factors. People whose estimated brain age came in lower than their actual chronological age experienced a lower risk.

A Potential Tool for Earlier Dementia Detection and Monitoring

Because EEG readings require no invasive medical procedures, the study authors suggest that sleep-based brain age assessments could eventually help evaluate dementia risk outside traditional clinical environments. Future wearable technologies might record the necessary brain signals while a patient sleeps.

“Brain age is calculated from sleep brain waves,” said Leng. “We know that brain activity during sleep provides a measurable window into how well the brain is aging.”

The results also point to the possibility that improving sleep health could influence how the brain ages. Leng noted that prior research demonstrates how treating sleep disorders can alter brain wave activity recorded during sleep. “Better body management, such as lowering body mass index and increasing exercise to reduce the likelihood of apnea, may have an impact,” said first author Haoqi Sun, PhD, an assistant professor of neurology at Beth Israel Deaconess Medical Center, who developed the model alongside co-authors Robert J. Thomas, MD, and M. Brandon Westover, MD, PhD. “But there’s no magic pill to improve brain health.”

Frequently Asked Questions

How does the machine-learning model calculate brain age?

The model combines 13 microscopic features found within electroencephalography (EEG) brain wave recordings collected while a person sleeps.

What is the link between estimated brain age and dementia risk?

Researchers found that for every 10-year increase between estimated brain age and actual chronological age, the likelihood of developing dementia rose by nearly 40 percent.

Did traditional sleep measurements predict dementia?

Who led the research study?

The research was led by scientists at UC San Francisco and Beth Israel Deaconess Medical Center in Boston, with findings published in JAMA Network Open.