BrainYears™ was developed in collaboration with the Buck Institute for Research on Aging — the world’s largest independent research institute focused solely on aging. The model was trained on data from over 650 participants aged 18 to 89, using 643 biomarkers derived from EEG, ERPs, and cognitive performance measures.
It was validated on a rigorous held-out test set (n=128) that the algorithm never saw during training. The model achieves a Pearson correlation of 0.92 and a mean absolute error of 4.4 years — matching MRI-derived brain age clocks in accuracy, from a wearable device. Accuracy is also stable across the adult lifespan.
The intervention study included 28 participants in the treatment group and 22 in a minimal-exposure comparison group. The validation study is published on BioRxiv and describes the full methodology.
Was this article helpful?
That’s Great!
Thank you for your feedback
Sorry! We couldn't be helpful
Thank you for your feedback
Feedback sent
We appreciate your effort and will try to fix the article