Software and Hardware Complex of Anthropomorphic Type Robot as an Assistant for a Teacher. Decision-Making Subsystem Using Multiscale Entropy Analysis of EEG Signals

Abstract

The paper presents research on the use of the results of the analysis of signals of brain activity of students for decisionmaking in educational robotics. Signals of brain activity were obtained using electroencephalograms (EEG). Assessment of the state of the student was carried out on the basis of the multiscale entropy. The object of the study was a man aged 17 years who was diagnosed with focal (structural) epilepsy, mesial sclerosis on the left and focal cortical dysplasia of the left temporal lobe and a control group. Comparison of the results of entropy estimates was carried out in the form of topographic images. Topographic images of the surface of the head are obtained on the basis of a spherical spline. The study showed that multiscale entropy of EEG signals can be a useful tool in the classification of patients with epilepsy and the control group. It is anticipated that such an analysis will be useful for early detection of neurological changes. The use of multiscale entropy in educated robotics as a means of obtaining objective information will help increase objectivity in decision-making in the choice of educational technologies to improve the quality of the educational process.