Comparison between neural network or neural network with genetic algorithm and analysis of EEG signal

Electroencephalogram (EEG) refers to the recording of the brain's spontaneous electrical activity. It consists of 5 sub-band signals that can be traced and analyzed to detect many diseases. The use of manual prediction cases is a difficult method to get an accurate classification of the EEG signal. This problem increases the number of misdiagnosis that commonly plagues all classification systems. This research aims to analyze Electroencephalogram (EEG) signal parameters using two ways: first, by using only the neural network; second, by using the Genetic Algorithm (GA) and Neural network and then comparing the two ways. A total of 80 well-known reference cases were used in this study. We had four cases of normal open eye, normal close eye, epilepsy free seizures and epilepsy seizures. Signals were classified based on statistical features and total power spectrum extracted from the signals. The feed-forward neural network was applied to classify the case of the signals.