Vonsevych K. Myographic system of bionic hand with optical identification of surface type

Vonsevych K. Myographic system of bionic hand with optical identification of surface type. – On the rights of manuscript. The dissertation for scientific degree of candidate of technical sciences on a specialty 05.11.17 – Biological and medical devices and systems. – National Technical University of Ukraine "Igor Sikorsky Kyiv Polytechnic Institute" MES of Ukraine, Kyiv, 2020. This work is devoted to creation of a myographic system of prosthetic hand with advanced movements and gestures capabilities with recognition of myoelectric signals by a neural network interface and optical identifier of the contact surface for fingers touch. The method of recognizing categories of physiological movements and gestures by analyzing electro- and force-myographic signals on the base of multilevel artificial neural networks have been improved, which allowed to increase the accuracy of classification of wrist gestures. The method of recognizing the contact surface by a finger prosthesis due to optical identification with instruments of concentration optical energy have been improved, which made possible to increase the reliability of identification of the structure of the objects of manipulation. The method of hand prosthesis movements coordination based on simultaneous registration and recognition of physiological signals and optical identification signal have been improved. Keywords: artificial neural network, bionic hand prosthesis, confusion matrix, force myography, optical identification, prosthetics, signal classification, surface electromyography, surface recognition, time domain features