Classification of Epileptic seizure EEG signals using EMD and ANFIS

Epileptic seizure by the Indian Epilepsy centre states that seizure is a commonest disorder of the brain where one out of 200 suffer from seizure. In this paper, the diagnostic significance of seizure EEG signals is investigated using empirical mode decomposition and neural network methods. The Epileptic seizure data of 25 subjects were obtained. Empirical mode decomposition analysis is carried out and the measure of intrinsic mode functions is used for feature extraction. The statistical features like interquartile range (IQR), minimum, maximum, variance and power of the signal is exracted. It is then used to train as well as to test the neural networks. The performance of the proposed approach, adaptive neuro fuzzy inference system (ANFIS) with back propagation artificial neural network is compared. The outcome of the result reveal that the ANFIS is more useful for categorizing seizure EEG signal. It becomes visible that the empirical mode decomposition and adaptive neuro fuzzy network is more perceptive and can be used for detecting epileptic seizures.