Rethinking Seizure Care Blog

Predicting Dementia in Older Adults Using Sleep Tests

Posted by RSC Diagnostics on Sep 3, 2021

Doctor holding clipboard with elderly lady lying in hospital bed

Dementia is a problem that affects many people as they get older. It’s estimated that there are at least 50 million people around the world. Unfortunately, there is no cure and it often doesn’t get diagnosed until people are in the later stages. However, there could be some good news when it comes to predicting dementia in older adults.

Researchers from Massachusetts General Hospital and Beth Israel Deaconess Medical Center have found and validated a marker for dementia. The researchers published their findings in JAMA Network Open in 2020.

The Brain Age Index

The group created the Brain Age Index (BAI). This is a model that makes use of artificial intelligence and a set of sleep data. It can then provide an estimate of the difference between the biological age of a person’s brain and that person’s chronological age. It can do this through the use of an electroencephalogram (EEG) when a person is sleeping.

If there is a higher BAI, it means that there is a change from the way that a brain normally ages. What does this mean? Researchers believe it has the potential to determine whether the person has dementia. It could also help to determine the severity of their dementia.

How do the researchers know if the BAI values are high? With the aid of AI, they took the values of thousands of sleep tests. These tests were from 88 people who have dementia, 44 people who had mild cognitive impairment, 1,075 with some cognitive symptoms but who were not diagnoses with impairment, and 2,336 who did not have dementia.

The research showed that the BAI values were higher in the groups that had cognitive impairment. Patients who had dementia were found to have a value that was about four years old than the patients who did not have dementia. The “BAI values also correlated with neuropsychiatric scores from standard cognitive assessments conducted by clinicians before or after the sleep study.”

How The BAI Could Help

Brandon Westover is an investigator at the Department of Neurology at Massachusetts General Hospital, as well as the director of Data Science at the MGH McCance Center for Brain Health. He said, “The model computes the difference between a person’s chronological age and how old their brain activity during sleep ‘looks,’ to indicate whether a person’s brain is aging faster than is normal.”

The first author of the study, Elissa Ye, is a member of Westover’s team. She said, “This is an important advancement because before now it has only been possible to measure brain age using brain imaging with magnetic resonance imaging, which is much more expensive, not easy to repeat, and impossible to measure at home.”

Easier Testing Can Lead to Better Care

According to Ye, the EEG tests are becoming more and more accessible even outside of the typical sleep lab environment. This is possible because the technology is better today. There are headbands and dry EEG electrodes that can be used. Because it is relatively easy to use, it could help to expand the usage of the tool.

It is possible to get data from EEGs over the course of multiple nights, even when patients are at home rather than in a sleep study lab. The researchers hope that eventually, the use of EEG for this type of prediction will become a normal part of primary care that patients can receive. It has the potential to help doctors get a better idea of which of their patients could suffer from dementia as they get older, allowing them to get proper treatments now and to prepare for their later life.



Topics: In-Home Video EEG