Classification of Ictal and Interictal Eeg Signals

An electroencephalogram (EEG) is a graphical record of ongoing electrical activity produced by firing of neurons of the human brain due to internal and/or external stimuli. Feature extraction and classification of the EEG signals are used for diagnosis the epileptic seizure (i.e., physical changes in behaviour that occur due to abnormal electrical activity in the brain). Classification of Ictal (i.e., seizure period) and Interictal (i.e., interval between seizures) EEG signals is very important for the treatment and precaution of an epileptic patient. However, the classification accuracy of Ictal and Interictal EEG signals is not at satisfactory level due to their non-abruptness phenomenon using the existing seizure and non-seizure classification methods. Moreover, the features of Ictal and Interictal signals are not consistence in different locations for an epileptic period. In this paper we present new approaches for features extraction of Ictal and Interictal including various transformations such as discrete cosine transform (DCT), DCT-discrete wavelet transform, and singular value decomposition. The least square support vector machine is applied on the features for classifications. Results demonstrate that our proposed methods outperform the existing state-of-the-art method in terms of classification accuracy for the large benchmark dataset in different brain locations.