AI can tell if your brain is aging faster than you are

A new machine-learning technique that scrutinizes brain activity during sleep may reveal individuals at increased risk of dementia, according to research led by scientists from UC San Francisco and Beth Israel Deaconess Medical Center in Boston. By analyzing electrical signals gathered through electroencephalography (EEG) while subjects sleep, the system estimates a person's brain age.

This brain age estimate reveals when the brain appears older than the person's actual chronological age. For each 10-year difference between estimated brain age and actual age, the likelihood of developing dementia rises by nearly 40%. Conversely, those with an estimated brain age younger than their actual age carry a lower dementia risk.

The findings, published in JAMA Network Open, were derived from a machine-learning model combining 13 microscopic features found within EEG brain wave recordings. This model was applied to data from approximately 7,000 participants in five separate studies. The participants, aged between 40 and 94 at the start of the studies and without dementia at the beginning, were monitored over periods ranging from 3.5 to 17 years.

During this time, around 1,000 participants developed dementia. The study found that subtle yet detailed patterns in sleeping brain waves, which traditional sleep measurements often miss, could provide insights that standard sleep assessments fail to detect. Senior author Yue Leng, MBBS, PhD, an associate professor of psychiatry at UCSF, explained that several EEG patterns used to determine brain age are already recognized for their role in memory and cognitive health.

For instance, delta waves, slow rolling electrical patterns characteristic of deep sleep, and sleep spindles, brief bursts of rapid brain activity that help consolidate memories, were among the features analyzed. The study also revealed that sudden spikes in EEG signals, known as kurtosis, were linked to a lower risk of dementia.

Even after accounting for factors such as education, smoking, body mass index, physical activity, other medical conditions, and genetic risk factors, the association between an older estimated brain age and higher dementia risk persisted. Leng emphasized that the brain activity during sleep offers a measurable window into how well the brain is aging.

Moreover, the findings suggest that enhancing sleep health might influence how the brain ages. The researchers believe that EEG readings, which can be collected non-invasively, could eventually serve as a tool for earlier dementia detection. Wearable technologies might one day be capable of recording the necessary brain signals during sleep.

The researchers noted that improving sleep health, including treating sleep disorders, managing body weight, and increasing exercise to reduce the likelihood of apnea, could potentially impact brain aging. However, they stressed that there is no guaranteed solution to improve brain health.