6th Edition of Neurology World Conference 2026

Abstract

Major depressive disorder is often characterized as a dynamic brain disease, reflecting the nonlinear alterations of the brain's electrical activity during the disease episodes. The instability in emotional processing results in a transition from a healthy behavior towards a pathological state, expressed by bifurcation from healthy brain dynamics to a pathological one. Different levels of EEG stationarity and nonlinearity may describe such alterations. Quantifying neural activity measured by EEG, however, requires approaches based on appropriate background.

The traditional analytical methods based on linear transformation often show controversial results due to the impropriety of capturing the features of brain electrical activity with nonlinear and irregular character. Techniques based on multiple time-frequency resolution, typical for wavelet analysis, allowed us to study changing levels of EEG stationarity during disease development. In addition, together with the methods based on nonlinear dynamics, such as the approximate entropy or fractal dimension, the neuropathological processes may be markedly recognized.

Depressive patients are repeatedly found to have a lower EEG complexity and higher predictability and regularity than controls expressed by the approximate entropy parameter, which corresponds to the level of new EEG signal pattern generation. Thus, EEG signals from depressive patients have been found to be more regular and predictable. Together with the other entropy measures, different degrees of EEG complexity were associated with various functional stages of the brain neuronal networks typical for the different stages of disease.

Such novel mathematical approaches represent great progress in diagnostic and prognostic procedures, providing the discrimination between normal and depression EEG signals with a high probability.