Epileptic electroencephalogram classification

Brain is very complex organ of the body. Electrical activity of the brain is referred as Electroencephalogram (EEG). EEG signal acts as a n important tool for diagnosis of neural diseases. The problem of EEG signal classification is a pattern recognition problem using extracted features. Observing EEG signals for seizure detection all times is a difficult task. So to make the detection easier, increase the accuracy and speed of analysis & classification it is necessary to invent different techniques. In this paper, to extract characteristics from given EEG data different methods of feature extraction have been applied. Features have been extracted by computing Fourier transform, discrete wavelet transform (DWT), empirical mode decomposition (EMD) & bispectrum analysis and seizure detection pattern is investigated for classification. The database which is publicly available at Bonn University is taken.