Multi-Stage, Multi-Resolution Method for Automatic Characterization of Epileptic Spikes in EEG

In this paper, a technique is proposed for the automatic detection of the spikes in long term 18 channel human electroencephalograms (EEG) with less number of data set. The scheme for detecting epileptic and non epileptic spikes in EEG is based on a multi resolution, multi-level analysis and Artificial Neural Network(ANN) approach. Wavelet Transform (WT) is a powerful tool for signal compression, recognition, restoration and multi-resolutionanalysis of non-stationary signal. The signal on each EEG channel is decomposed into six sub bands using a non-decimated WT. Each sub band is analyzed by using a non-linear energy operator, in order to detect spikes. A parameter extraction stage extracts theparameters of the detected spikes that can be given as the input to ANN classifier. A robust system that combines multiple signal-processing methods in a multistage scheme, integratingwavelet transform and artificial neural network is proposed here. This system is experimented on a simulated EEG pattern waveform as well as with real patient data. The system is evaluated on testing data from 81 patients, totaling more than 800 hours of recordings.90.0% of the epileptic events were correctly detected and the detection rate of non epileptic events was 98.0%. We conclude that the proposed system has good performance in detecting epileptic form activities; further the multistage multiresolution approach is an appropriate way of automatic classification problems in EEG.