Automatic detection of epileptic seizures using Independent Component Analysis Algorithm

Epilepsy is a disorder in which the brain cells release abnormal electrical signals. The method presented here is for the detection of epileptic seizures from background EEG. Two techniques namely Principle Component Analysis (PCA) and Independent Component Analysis (ICA) are applied for the epileptic spike detection. PCA is not able to separate the epileptic spikes. It is found that ICA performs better in detection of epileptic spikes. ICA is performed on the EEG data with the epileptic seizures and based on the Hurst exponent (H) value the spikes corresponding to epileptic seizures are detected. The wavelet transform technique is attempted to increase the detection rate in comparison to the threshold technique.