关键词: 脑肌电耦合/
变分模态分解/
传递熵/
时频尺度
English Abstract
Functional coupling analyses of electroencephalogram and electromyogram based on variational mode decomposition-transfer entropy
Xie Ping1,Yang Fang-Mei1,
Li Xin-Xin1,
Yang Yong1,
Chen Xiao-Ling1,
Zhang Li-Tai2
1.Key Laboratory of Measurement Technology and Instrumentation of Hebei Province, Institute of Electric Engineering, Yanshan University, Qinhuangdao 066004, China;
2.Department of Rehabilitation Medicine, the No. 281 Hospital of Chinese People's Liberation Army, Qinhuangdao 066100, China
Fund Project:Project supported by the National Natural Science Foundation of China (Grant No. 61271142), and the Natural Science Foundation of Hebei Province, China (Grant Nos. F2015203372, F2014203246).Received Date:26 January 2016
Accepted Date:02 March 2016
Published Online:05 June 2016
Abstract:The functional corticomuscular coupling (FCMC) is defined as the interaction, coherence and time synchronism between cerebral cortex and muscle tissue, which could be revealed by the synchronization analyses of electroencephalogram (EEG) and electromyogram (EMG) firing in a target muscle. The FCMC analysis is an effective method to describe the information transfer and interaction in neuromuscular pathways. Forthermore, the multiscaled coherence analyses of EEG and EMG signals recorded simultaneously could describe the multiple spatial and temporal functional connection characteristics of FCMC, which could be helpful for understanding the multiple spatial and temporal coupling mechanism of neuromuscular system. In this paper, based on the adaptively decomposing signal into frequency band characteristis of variational mode decomposition (VMD) and the quantitatively detecting the directed exchange of information between two systems of transfer entropy (TE), a new methodvariational mode decomposition-transfer entropy (VMD-TE) is proposed. The VMD-TE method could quantitatively analyze the nonlinear functional connection characteristic on multiple time-frequency scales between EEG over brain scalp and surface EMG signals from flexor digitorum surerficialis, which are recorded simultaneously during grip task with steady-state force output.In this paper, application of VMD-TE method consists of two steps. Firstly, the EEG and EMG signals are adaptively decomposed into multi intrinsic mode functions based on variational mode decomposition method, respectively, to describe the information on different time-frequency scales. Then the transfer entropies between the different timefrequency scales of EEG and EMG are calculated to describe the nonlinear corticomuscular coupling characteristic in different pathways (EEGEMG and EMGEEG), to show the functional coupling strength (namely VMD-TE values). finally, the maximum VMD-TE values between the different time-frequency scales of EEG and EMG signals among the eight subjects are selected, to describe the discrepancies of FCMC interaction strength between all time-frequency scales. The results show that functional corticomuscular coupling is significant in both descending (EEGEMG) and ascending (EMGEEG) directions in the beta-band (15-35 Hz) in the static force output stage. Meanwhile, the interaction strength between EEG signal and the gamma band (50-72 Hz) of EMG signal in descending direction is higher than in ascending direction. Our study confirms that the beta oscillations of EEG travel bidirectionally between sensorimotor
Keywords: functional coupling/
ariational mode decomposition/
ransfer entropy/
ime-frequency scales
