Localization of epileptic seizure with an approach based on the PSD with an autoregressive model

摘要: In this study, we present a criterion based on the analysis of EEG signals through mean conventional power spectral density (PSD) in aim to localize and detect epileptic area brain. Firstly, as are commonly non stationary practice, processed data with technique differentiation order have which is convenient model autoregressive (AR). For this, used many techniques for determine better work. Therefore, can characterize normal abnormal activity correspond discharge patient. Our contribution work automatic detection epilepsy seizure PSD novel approach by resolution frequency domain examination often done visual inspection rhythm (delta, theta, alpha, beta, gamma) neurologists practitioners. The accuracy estimated 70% sensitivity 80.55% compared interpretation neurologist.