Application of a variation of empirical mode decomposition and teager energy operator to EEG signals for mental task classification

This paper presents a simple and effective methodology for mental task classification using a novel variation of the empirical mode decomposition (EMD) algorithm and the Teager energy operator applied to electroencephalography (EEG) signals. EEG signals corresponding to various types of mental tasks performed by human subjects are decomposed using the variation of EMD, called Empirical Mode Decomposition-Modified Peak Selection (EMD-MPS), which allows direct separation of the signals into a de-trended component, and a trend, according to a frequency separation criterion. Teager energy operator is then applied to calculate the average energy values of both components obtained after signal decomposition using EMD-MPS. These energy values are used to construct feature vectors, and one-versus-one classification of mental tasks is performed using a simple classifier, namely the 1-NN classifier. An average correct classification rate of 87% is obtained, improving on previous results and thereby also demonstrating the effectiveness of the methodology.