Research On Intermuscular Coupling Analysis Method Of Upper Limb Movement

Stroke is a localized or diffuse cerebral dysfunction on the basis of cerebrovascular disease or blood flow disorder,which has become the main cause of adult acquired dyskinesia.Some studies have shown that it is precisely because of the damage of upper motor neurons and the loss of control of motor system in stroke patients,which leads to partial or complete blocking of motor control and sensory nerve pathway,which leads to abnormal motor mode.How to promote the rehabilitation of upper limb motor function in stroke patients is a hotspots and difficulties of rehabilitation medicine at the moment.Intermuscular coupling(IMC)refers to the correlation and interaction between different muscles under the control of the motor central nervous system(CNS),which can reflect the functional connections between the cerebral cortex and muscles at different levels.Therefore,the study of IMC characteristics can explore the mechanism of motor dysfunction in stroke patients,which is expected to provide theoretical and physiological evidence for the course of assessment,training of rehabilitation,which is of great significance.First of all,this paper introduces the generation and characteristics of electromyography(EMG)signals,summarizes the research progress of IMC analysis,sorts out several major public EMG data sets,and compares different types of IMC analysis algorithms based on the IMC relationship.In view of the limitations of the existing methods,the main research content of this paper is determined: the theory of Copula in statistics is introduced into the analysis of IMC,which provides a brand-new inspiration to understand the complex phenomenon of IMC.Secondly,in view of the shortcomings of existing IMC analysis methods,such as coherence and mutual information,which can only describe the static coupling strength,and cannot accurately calculate the linear and nonlinear coupling strength,a new time-varying Copula mutual information estimation method is proposed and applied to the surface EMG(s EMG)signals recorded by biceps brachii(BB)and triceps brachii(TB)during wrist flexion and wrist abduction coupling analysis in different characteristic frequency bands.The experimental results show that the time-varying Copula function has higher goodness-of-fit than the intermuscular dependent structure characterized by static Copula function.There are significant differences in the frequency band of the IMC strength derived from the time-varying Copula mutual information.The specific performance is that the higher the frequency band,the lower the IMC strength,while the static Copula mutual information wrongly underestimates IMC strength.Thirdly,in view of existing IMC analysis methods,such as Granger causality and transfer entropy,which are not enough to accurately identify the direction of intermuscular causality information,a new Gaussian Copula transfer entropy estimation method is proposed.Based on this foundation,the intermuscular causal coupling network models of different frequency bands are constructed,and the intrinsic properties of IMC are analyzed by network topological parameters.The experimental results show that the IMC is bidirectional,and that in different frequency bands is the same but different.The middle deltoid(MD),posterior deltoid(PD)and TB are the important nodes in the IMC network,which share similar coupling relationship.However,BB is abnormal,which is non-critical node in the IMC network.There was no significant difference in the IMC network between stroke patients with mild motor dysfunction and healthy people.Finally,in view of the lack of effective description of direct and indirect nonlinear coupling between multi-channel signals,a new estimation method of R-Vine Copula mutual information and R-Vine Copula conditional mutual information is proposed,which can analyze the characteristics of IMC under a unified theoretical framework.The experimental results show that the probability density function of s EMG signal may conform to peak thick tail distribution.Compared with other Vine Copula functions,R-Vine Copula function is more flexible and reasonable in describing the intermuscular dependent structure.The binary t Copula function is probably the most suitable Pair-Copula function.There is a high coupling strength between TB and the MD,while the degree of coupling between BB,pectoralis major(PM)and other muscles is lower,and the IMC between subjects has significant correlation.