Seizure detection by means of Hidden Markov Model and Stationary Wavelet Transform of electroencephalograph signals

In this paper, the intracranial Electroencephalograph (EEG) has been employed to detect and classify the states of seizures in human subjects. In so doing, the Stationary Wavelet Transform (SWT) has been deployed to extract features from EEG signals recorded from several patients suffering from medically intractable focal epilepsy. All together three states of seizures have been considered: (i) ictal, (ii) preictal, and (iii) interictal. The classification is achieved by means of the Hidden Markov Model (HMM). It will be shown in this paper that the methods and procedures can accurately predict epileptic seizure patterns with both sensitivity and specificity of more than 96%.