HYBRID MODELING OF EEG: THE FITZHUGH-NAGUMO-LORENZ MODEL

The paper presents a method of modeling electroencephalographic (EEG) signals using a hybrid biophysical model that combines the FitzHugh-Nagumo dynamics and the chaotic Lorentz system. A comprehensive approach to optimizing the model parameters based on neural networks is developed, which automatically adjusts the parameters to maximize the fit to real EEG data. The proposed model demonstrates the ability to reproduce the charac-teristic features of EEG signals, including the main rhythms and the corresponding spectral characteristics. An interactive software tool developed in MATLAB provides a convenient in-terface for use. The results demonstrate the potential of this approach for both neuroscientific research and clinical applications in the diagnosis and modeling of pathologies.