Overview
A machine learning approach applied to electroencephalogram (EEG) recordings indicated that an individual's sleeping brain activity can reveal indicators of dementia risk prior to the manifestation of memory impairments. The analysis, conducted on data from approximately 7,000 adults, established a correlation between an estimated "brain age" that exceeded chronological age and a heightened predisposition to dementia. This research identified that for every increment of 10 years in estimated brain aging beyond an individual's actual age, the associated risk of developing dementia increased by nearly 40%.
Approach
Researchers employed machine learning techniques to process and interpret EEG recordings. These recordings were sourced from approximately 7,000 adult participants. The analytical focus was on determining a "brain age" from these EEG data. The methodology involved comparing this derived brain age against the chronological age of the individuals to identify discrepancies, specifically where the estimated brain age was greater than the individual's actual age.
Findings
The study's findings directly linked an older-than-expected "brain age" to a substantially elevated risk of dementia. Specifically, the analysis revealed a quantitative relationship: for each additional 10 years that an individual's estimated brain age surpassed their chronological age, the associated risk of developing dementia increased by nearly 40%.
Why This Matters
The observed capacity for a person's sleeping brain to exhibit potential warning signs of dementia, as detected through EEG analysis and machine learning, suggests the possibility of identifying individuals at increased risk well in advance of the onset of memory problems.