Artificial intelligence systems for classifying EEG responses to imaginary and real movements of operators

摘要

Here, we introduce the method based on artificial neural networks (ANNs) for recognition and classification of patternsin electroencephalograms (EEGs) associated with imaginary and real movements of untrained volunteers. In order to getthe fastest and the most accurate classification performance of multichannel motor imagery EEG-patterns, we proposeour approach to selection of appropriate type, topology, learning algorithm and other parameters of neural network. Weconsidered linear neural network, multilayer perceptron, radial basis function network (RBFN) and support vectormachine. We revealed that appropriate quality of recognition can be obtained by using particular groups of electrodesaccording to extended international 10−10 system. Besides, pre-processing of EEGs by low-pass filter can significantlyincrease the classification performance. We developed mathematical model based on ANN for classification of EEGpatternscorresponding to imaginary or real movements, which demonstrated high efficiency for untrained subjects.Achieved recognition accuracy of movements was up to 90−95% for group of subjects. RBFN demonstrated moreaccurate classification performance in both cases. Pre-filtering of input data using low-pass filter significantly increasesrecognition accuracy on 10−20% in average, and the low-pass filter with cutoff frequency 4 Hz shows the best results. Itwas revealed that using different sets of electrodes placed on different brain areas and consisted of 6-12 channels, onecan achieve close to maximal classification accuracy. It is convenient to use electrodes on frontal and temporal lobes forreal movements, and several sets containing 6-9 electrodes — in case with imagery movements.