The use of wavelet transform as a preprocessor for the neural network detection of EEG spikes

Abstract

In this study, the wavelet transform is used to process EEG data as input to a feed forward neural network for the detection of epileptogenic transient waveforms. The compression capability of wavelet transform provided the inclusion of data before and after the spike for contextual information without increasing input size of the neural network. The network is trained for the detection of spikes and non-spikes. The results show that wavelet transform can be used to provide more relevant information for improving the detection of epileptogenic spikes for automated EEG monitoring of seizure patients.<>