Research On Design And Evaluation Method Of Hand Function Rehabilitation System Based On SEMG

Stroke is a disease that poses a great threat to human hand motor function.More than80% of stroke patients have different degrees of hand motor dysfunction.In different periods of illness,motor dysfunction is different,and the existing rehabilitation training system is often only suitable for a certain stage of rehabilitation training.The rehabilitation hand used has the problems of too large volume,poor wearability and high cost.Rehabilitation begins with evaluation and ends with evaluation.When formulating the rehabilitation plan of hand motor function,it is necessary to accurately evaluate the function of the opponent.However,most of the evaluation methods used in clinic today are semi quantitative evaluation scales,which lack objectivity,can not obtain accurate human physiological information,and can not understand the state of human motor function,so it is difficult to formulate the best treatment plan.Therefore,how to design a stroke hand function evaluation method is particularly important.Surface electromyography signal(sEMG)is a bioelectric signal produced by human body that can characterize the degree of muscle activity.Can sEMG be used to assist stroke patients in rehabilitation training and evaluation.Aiming at the above problems,this paper designs a hand function rehabilitation system based on sEMG,studies the hand function evaluation method,and puts forward the method of using sEMG to evaluate the hand function of stroke patients.The specific research contents are as follows:Aiming at the urgent need of a perfect hand function rehabilitation system for stroke patients,this paper designs a hand function rehabilitation system based on sEMG.The rehabilitation system selects appropriate sEMG acquisition equipment to collect sEMG,and puts forward three different training methods: passive training,mirror training and active training by analyzing the hand function characteristics of stroke patients in each rehabilitation period and the rehabilitation needs of stroke patients in different periods.According to the characteristics of each rehabilitation training mode,the hardware and software of the system are designed.Due to the defects of the existing rehabilitation equipment,this paper designs an exoskeleton rehabilitation hand assisted rehabilitation training for stroke patients.Analyze the hand function characteristics of patients with different grades in Brunnstrom grade evaluation method and Fugl Meyer scale evaluation method,and design rehabilitation training actions combined with expert opinions.Analyze the physiological characteristics and related parameters of the hand,and complete the mechanical structure design of the exoskeleton rehabilitation hand in combination with the requirements of rehabilitation training.Finally,complete the relevant circuit and software design of the exoskeleton rehabilitation hand,and verify and optimize the function of the designed exoskeleton rehabilitation hand.Experiments show that the hand function evaluation method based on sEMG proposed in this paper is fully feasible.This paper uses the sEMG data of 50 stroke patients for experimental verification,selects four movements of fist clenching,hand opening,wrist flexion and wrist extension as hand function evaluation actions,and collects the sEMG when the patients complete the above four movements for SVM,BP neural network and LVQ neural network training.The experimental results show that the above four algorithms fail to meet the expected requirements.Therefore,based on the optimization of LVQ neural network,this paper proposes a hand function evaluation model of stroke patients combined with GLVQ neural network and time-domain characteristics,which can improve the classification accuracy of five Brunnstrom levels to 98%.