Diagnosis of Epilepsy from EEG signals using Hilbert Huang
In this paper application of Hilbert-Huang transform (HHT) to electroencephalogram (EEG) signals analysis is presented, in order to simplify diagnosis of epilepsy. HHT consists of empirical mode decomposition (EMD) and Hilbert spectral analysis. Hilbert marginal spectrum represents a contribution of total amplitude (or energy) over various frequency values. This approach is used for analyzing 200 EEG signals, where half of all signals are from healthy subjects and the other half are signals of subjects with epileptic syndrome without seizure. It is showed that this method provides clear distinctions in visualisation of EEG signals of healthy and ill subjects, so that it can efficiently identificate epilepsy syndrome. Further, this approach can be a foundation for development of simple automated EEG signal classifiers in the aspect of recognition subjects with epilepsy syndrome and find its place in standard clinical practice. Keywords : EEG, epilepsy, Hilbert Huang transform, Hilbert marginal spectrum