Do try this at home: Age prediction from sleep and meditation with large-scale low-cost mobile EEG
Imaging Neuroscience January 1, 2024 Hubert Banville, Maurice Abou Jaoude, Sean U. N. Wood et al. 7 citations
Using a portable, four-channel consumer EEG device, age can be predicted from brain activity recorded during at-home meditation and sleep. Analyzing data from over 5,200 people aged 18–81, machine-learning models predicted chronological age with cross-validated R² scores between 0.3 and 0.5, matching the accuracy of lab-grade EEG benchmarks. Sleep recordings outperformed meditation recordings; the N2 and N3 sleep stages contributed most to predictions, but combining all sleep stages yielded the best performance. Age-related information was distributed across electrodes and frequencies, favoring multivariate models. Longitudinal data from eight subjects showed that EEG-based age predictions reflect both stable traits and day-to-day fluctuations.