Multinational qEEG developmental surfaces

Abstract

AbstractThe quantitative electroencephalogram (qEEG) is a diagnostic method based on the spectral features of the resting state EEG. The departure of spectral features from normality is gauged by the z transform with respect to the age-adjusted mean and deviation of normative databases – known as the developmental equations/surfaces. However, the extent to which the data collected from different countries with various equipment require separate developmental equations remains unanswered. Here, we analyzed the EEG of 535 subjects from 3 countries, Switzerland, the USA and Cuba. The EEG power spectra of all samples were log transformed and their relations to the covariables (‘age’, ‘frequency’, ‘country’ and ‘individual’) were analyzed using the linear mixed effects model. We found that the origin ‘country’ of the subjects did not play a significant effect on the log spectra, even without interactions with other independent variables, whereas, ‘age’ and ‘frequency’ were highly significant. To estimate the developmental surfaces in greater detail, we carried out kernel regression (lowess) in two dimensions of log-age and frequency. We found two main phenomena: 1) slow rhythms (δ, θ) predominated in the lower ages and then decreased with a tendency to disappear at higher ages; 2) α rhythm was absent at lower ages, but gradually appeared more relevant in occipital and parietal regions, and increased with aging with an increasing centering frequency of α rhythm. We consider both phenomena as an expression of healthy neurodevelopmental and maturation related to age. It is the first study of multinational qEEG developmental surfaces accounting for ‘country’. The results demonstrate the possibility of creating international qEEG norms since the ‘individual’ and ‘age’ variability are much larger than the specific factors like ‘country’, or the technology employed ‘device’.