The early detection and quantification of abnormal brain changes are important for the prospective identification and subsequent treatment of individuals at risk for various neurological disorders, injuries, and cognitive decline. Discrepancies between an individual’s chronological age and his or her neuroimaging-based “brain age” can reveal a “brain-age gap” that is a biomarker for a range of neurological disorders, including Alzheimer’s disease and other forms of dementia. Previous techniques for brain-age estimation unfortunately require expensive MRI scanning, which limits their practical use.

Addressing this issue, researchers at Drexel have developed and validated a new robust, precise, reliable, and inexpensive EEG-based machine-learning technique for accurately assessing whether an individual’s brain is aging more quickly or more slowly than is typical for healthy individuals.  Based on our large database of EEGs, our results show that brain-age estimation with low-cost EEG recordings can be performed with a precision comparable to, or better than, estimations obtained with MRI, thus providing a breakthrough technology for studying and screening for the effects of disease, injury, genetics, and lifestyle factors such as exercise, nutrition, stress, and sleep on brain aging. We have also produced a user-friendly software package that applies our machine-learning technology to estimate an individual’s brain age.  Additionally, for a user who wishes to compare individuals to a different baseline group (e.g., soldiers, athletes, or students), the software can compute a new machine-learning brain-age model based on EEGs from that group.

Our innovative technology expands the practical use of brain-age estimation because EEGs can be easily and inexpensively recorded in the field, including educational and medical institutions. Consumer-grade EEG units can also be deployed for home use.