Abstract
Purpose of research . Robotic rehabilitation systems for patients with motor impairments are currently widely used. The main trend is to eliminate the exoskeleton operator from the process and transfer rehabilitation control functions to software. However, currently available exoskeleton control software and algorithms do not adequately synchronize the patient's "movement intentions" with the exoskeleton's assistive capabilities, resulting in reduced rehabilitation effectiveness. The aim of the study is to improve rehabilitation effectiveness by adapting exoskeleton control modes to the patient's psychophysiological and physical condition. Methods . A biotechnical system has been proposed for the rehabilitation of post-stroke patients with impaired lower extremity motor function. This system, using decoders of real and imagined movements based on the analysis of electroencephalographic and electromyographic signals, controls rehabilitation modes determined by the patient's current functional state. A method for generating a "Movement Intention" marker is proposed. This method is based on a comprehensive analysis of three electrophysiological signals generated in response to a stimulus included in the biofeedback system. This method allows for synchronizing exoskeleton control with the patient's imagined and real muscle contractions through biotechnical and biological feedback. Results. The effectiveness of the personalized rehabilitation method was assessed in the experimental and control groups. The Berg Balance Scale was used to assess rehabilitation effectiveness. Statistical analysis of score chang es in the experimental and control groups, based on the rehabilitation course, showed that the change in scores on the Berg Balance Scale exhibited a significant statistical difference when comparing the two study groups. Conclusion . The data obtained from the study demonstrate the effectiveness of using a personalized biotechnical system for restoring lower limb motor function in patients who have suffered an ischemic stroke

This publication has 12 references indexed in Scilit: