CODE IMAGES OF ELECTRIC CELL INFORMATION SIGNAL SIGNALS FOR CONTROLLING ROBOT-TECHNICAL DEVICES BY MEANS OF BRAIN-COMPUTER INTERFACE Federal State Educational Institution of Higher Educati

A method based on the use of code images obtained by generating a set of code messages on a certain EEG segment is proposed for decoding EEG in brain-computer interfaces. A code message is generated by encoding EEG signals at the outputs of a block of band-pass filters. In the frequency range of the EEG, four frequency bands are allocated, which corresponds to four channels for each EEG lead. The code messages of the four channels form the image of the EEG, which, when decoded, receives control signals to the servos of the robotic device. The image of the code messages is formed on the basis of the theory of multisets. For training the EEG image classifier, a software and hardware complex is used, including an electromyograph, an electroencephalograph, a band-pass filter unit and a computing device that discrete the signals from the electromyograph output and the band pass filter unit. The label of the image class was determined by the electromyograph lead signal corresponding to the motor unit being classified. Records with the fields of the code image and the corresponding class label of the control command are placed in the database. The proposed method is an alternative to the method of EEG decoding based on biofeedback.

1. Gerardin Е., Sirigu A., Leherici S. et al. Partially overlapping neural networks for real and imagined hand movements. Cerebral Cortex. 2000. V. 10. P. 1093.

2. Chernikova L.A., Mokienko O.A., Roshchin V.Yu., Bobrov P.D., Frolov A.A. Sposob reabilitatsii bol'nykh, perenesshikh insul't. Patent RU 2523349.

3. Kotov S.V., Turbina L.G., Bobrov P.D., Frolov A.A., Pavlova O.G., Kurganskaya M.E., Biryukova E.V. Primenenie kompleksa «interfeys “mozgkomp’yuter” i ekzoskelet» i tekhniki voobrazheniya dvizheniya dlya reabilitatsii posle insul'ta // Al'manakh klinicheskoy meditsiny. – 2015. – No. 39. – pp. 15-21.

4. Sotnikov P.I. Obzor metodov obrabotki signala elektroentsefalogrammy v interfeysakh mozg-komp’yuter // Inzhenernyy vestnik. MGTU im. N.E. Baumana. Elektronnyy nauchno-tekhnicheskiy zhurnal. – 2014. – No. 10. – pp. 621-632.

5. Rybochkin A.F., Savel'ev C.V., Pleskanos L.V. Analiz fotopletizmogrammy s ispol'zovaniem kodovykh soobshcheniy//Biomeditsinskaya radioelektronika No.2 2010. pp.43-48

6. Сотников П.И. Выбор оптимальных частотных диапазонов сигнала электроэнцефалограммы в интерфейсе мозг-компьютер // Наука и Образование. МГТУ им. Н.Э. Баумана. Электрон. журн. – 2015. – № 06. – С. 217-234.

7. Pyatin V.F., Kolsanov A.V., Sergeeva M.S. i dr. Informatsionnye vozmozhnosti ispol’zovaniya myu- i beta-ritmov EEG dominantnogo polushariya v konstruirovanii neyrokomp’yuternogo interfeysa // Fundamental'nye issledovaniya. – 2015. – No. 2-5. – pp. 975-978.

8. Stankevich L.A., Son'kin K.M., Shemyakina N.V. i dr. Klassifikatsiya EEGpatternov voobrazhaemykh dvizheniy pal’tsami odnoy ruki, vypolnyaemykh v zadannom ritme // Fiziologiya cheloveka. – 2016. – Vol. 42. – No. 1. pp. 40- 52.

9. Kudryavtsev, P.S. Razvitie metodologii bustinga dlya klassifikatsii flyuorogramm grudnoy kletki / P.S. Kudryavtsev, A.A. Kuz'min, S.A. Filist// Biomeditsinskaya radioelektronika. – 2016. – No.9. – pp. 10-14.

10. Filist S.A., Shatalova O.V., Efremov M.A. Gibridnaya neyronnaya set' s makrosloyami dlya meditsinskikh prilozheniy//Neyrokomp'yutery. Razrabotka i primenenie, 2014. - No.6. - pp. 35-39.