umtual information based feature selection for seizure detection in newborns EEG signals

a new automated method is proposed to detect seizure events in newborns from electroen cephalogram EEG data the detection scheme is based on observing the changing behavior of the wavelet coeffi cients WCs of the EEG signal at different scales an optimal feature subset is obtained usngi the mutual information evaluation function MIEF the MIEF algorithm evaluates a set of candidate features extracted from the WCs to select an informative feature subset the subset is then fed to an artificial neural network ANN calssifier that organizes the EEG signal into seizure or non-seizure activities.

Keywords:

EEG , seizure detection , newborn , discrete wavelet transorm feature extraction artificial neural network , optimization , mutual information evaluation function ,