Classification of Raw Electroencephalogram Signals for Diagnosis of Epilepsy Using Functional Connectivity

Abstract

Epilepsy diagnosis depends on the evaluation of long-term electroencephalogram (EEG) recordings performed by trained professionals, turning it in a time-consuming process that is not readily available for most patients. Furthermore, raw EEG recordings are inherently noisy, which makes EEG analysis troublesome. Thus, the present work proposes a methodology for automatic EEG classification of epileptic subjects which uses raw short-duration EEG recordings obtained with the patient at rest. The system is based on machine learning algorithms that use an attribute extracted from the power spectral density of EEG signals. This attribute is an estimate of functional connectivity between EEG channel pairs and is called debiased weighted phase-lag index (dWPLI). The classification algorithms we used were support vector machines (SVM) and artificial neural network (ANN). EEG signals were acquired during the interictal state, i.e., between seizures, and had no epileptiform activity. Both raw and pre-processed recordings of 15 epileptic patients and 10 healthy subjects were used to evaluate the method’s performance. The algorithms reached their maximum classification performances, 100% accuracy, using input attributes extracted from raw data. The results show the feasibility of the proposed system, given its high classification performance.