The Experimental Research About The Effects Of Rehabilitation Patterns For Upper Limb On Cortical Activities In CNS Circuits

Stroke patients often have upper extremity motor dysfunction,scientific and reasonable rehabilitation can help restore their motor function,improve their self-care ability and quality of life.In recent years,based on the neuroplasticity theory and the function mirror neuron system theory,people have developed different rehabilitation techniques,such as Neuromuscular Electrical Stimulation(NMES)and Virtual Reality(VR).NMES is a peripheral intervention technology that regulates neural remodeling through peripheral electrical stimulation to modulate damaged cortex;VR is a central intervention technology that directly stimulates the mirror neuron system through visual feedback to promote functional reorganization.No matter what kind of rehabilitation technology,it is through the activation of damaged sensorimotor cortex to promote cortical remodeling and mirror neuron regeneration.However,the traditional NMES focuses on the regulation of target muscle motor neurons,but ignores the important role of sensory information on cortical excitement and remodeling.And,at present,there are few studies that combine the two major rehabilitation theories to improve the patient’s upper limb motor function rehabilitation effect.Currently,Event-related Desynchronization(ERD)is commonly used to quantify the level of cortical excitement in the sensorimotor cortex.For this reason,this thesis designed the EEG acquisition experiment of the elbow joint motion evoked by NMES and the EEG acquisition experiment of the elbow joint movement guided by the motion visualization of the virtual upper limb platform.First,the electrical stimulation of "peripheral intervention" activates the cortical response by the "peripheral-central" pathway to explore a better NMES pattern,thereby more effectively promoting neural remodeling in the sensorimotor cortex.In this thesis,the EEG acquisition of sensory motor cortex induced by different NMES modes was designed,and NMES parameters were compared to modulate the excitability of sensorimotor cortex.Due to experimental requirements,an experimental platform was developed based on the NMES device,EEG acquisition device,elbow kinematics parameter acquisition device,and computer.The NMES models of Low-frequency pulsed Current(PC)and Kilohertz Frequency Alternating Current(KFAC)were designed to use for evoked elbow movement experiments.The experimental recruited 11 subjects for active elbow movement,passive elbow movement,and NMES-evoked elbow movement,and collecting the EEG signals in sensorimotor cortex and elbow kinematic parameters.In order to facilitate the rapid and batch processing of EEG signals,this thesis designed a software for EEG signal preprocessing,it was based on Independent Component Correlation Algorithm(ICA)algorithm,and Event-related Spectral Perturbation(ERSP)under the MATLAB software platform,this software was used to extract ERD features.The statistical results of ERD eigenvalues showed that the KFAC electrical stimulation model was generally stronger than the PC model in inducing sensorimotor cortical responses,which may be a better electrical pattern for the functional remodeling of damaged brain regions in stroke patients.The stimulation method is more conducive to the rehabilitation of the upper extremities’ motor function.Then,using the "central intervention",that is,the elbow joint motion in the virtual reality feedback environment causes the sensorimotor cortex response to activate the mirror neuron system,in order to investigate the upper extremity motor rehabilitation therapy that directly stimulates the "central" based on virtual reality technology.This paper designed an EEG acquisition experiment under active control of virtual upper elbow flexion,observation of virtual upper elbow flexion and guiding elbow flexion of virtual upper elbow flexion to to study the effects of different virtual reality training methods on sensorimotor cortical excitability.For this reason,this thesis used the virtual reality toolbox(Simulink)in MATLAB 8.4 to build a virtual upper limb platform,completed the real-time collection and processing of joint motion parameters on limb,modeling the virtual upper limb,scenario design of virtual upper limb platform,and control model design of the virtual upper limb movement,et al.,which achieved real-time interactive control between the user and the virtual upper limb model through the inertial sensor.The EEG data of 9 subjects were collected in the experiment,and then the ERD eigenvalues were calculated using the EEG signal processing software developed in this thesis.The results showed that the movement observation,active execution and action-guided execution based on the virtual reality scene can all induce sensorimotor cortical responses,in which the virtual upper limb movement guidance task has the best ERD response,indicating its cortical reorganization potential for promoting upper extremity motor function in patients with stroke.The aim of those studies is to explore that which NMES from "peripheral intervention" can help patients with upper extremity motor rehabilitation and use of virtual reality technology from "central intervention" to activate mirror neurons to help patients undergo upper limb rehabilitation training,and proposed an optimized NMES model.A virtual upper limb platform was set up,and a rehabilitation training model based on a virtual upper limb platform was designed,which provides a technical basis for the realization of closed-loop rehabilitation therapy combining "peripheral intervention" and "central intervention" in the future.