双间三模态神经肌肉活动范式下的脑—肌电信号处理与相干性分析

发布时间：2018-04-09 10:09

本文选题：神经肌肉活动　切入点：正中神经　出处：《天津大学》2012年硕士论文

【摘要】：人体的神经肌肉活动伴随着相应电生理信号的释放。近年来,运用脑电与肌电信号的协同分析考察神经肌肉活动状态成为运动医学、康复工程等领域的研究热点。典型的神经肌肉活动范式包括自主动作模态、刺激动作模态以及想象动作模态等,这些动作因脑肌控制机理的不同,其诱发出的脑电与肌电信号也存在着模式上的差异,通过比较分析多动作模态下脑电与肌电信号的内在联系规律,可以更深入地理解运动与大脑间的关系。本文以屈指这一神经肌肉活动范式为例,设计了自主、正中神经刺激和想象三种动作模态,并根据动作的时间间隔设定长间(相邻动作时间间隔10s)和短间(相邻动作时间间隔2s)两类实验方案,同步采集C3/C4导联处的头皮脑电和指浅屈肌处的表面肌电信号,经两点波峰阈值检测等预处理后,使用基于Morlet小波的时频分析方法初步研究了各模态肌电与脑电的独立变化规律。研究中利用偏定向相干算法、改进互相关系数极值点及时频特征分析对脑-肌电信号在不同动作模态和不同时间间隔下的相干性进行了比较与分析。文中研究结果表明:自主动作模态下,对侧脑电对肌电信号存在明显的支配作用,并与同侧脑电存在信息交互的现象;刺激动作模态下,肌电则成为同侧和对侧脑电的共同信源,且对对侧脑电的影响高于同侧;想象动作模态下,脑电和肌电信号无明显的因果性,但同侧与对侧脑电存在信息交互。同样的动作模态下,短间的偏定向相干要显著高于长间,而在同样的时间间隔下,刺激动作的因果性要显著高于自主动作,而想象动作为最低。短间和长间两大类动作中均会发生β段脑-肌电相干性极值点集中的现象,且短间相干性极值点数显著高于长间。时频特征分析结果也表明短间和长间存在一定差异,初步揭示了时间间隔对相干性的影响。本文通过不同时间间隔和不同动作模态下的脑肌电信号的联合分析,发现短间和刺激模态下的脑肌电存在高因果性,可能与短间动作下脑电信息产生交叠、刺激模态下大脑神经元更活跃且对肢体动作有较强的支配性有关。研究成果有助于更深刻的认识人体神经肌肉系统的运动能力和协调能力,未来有望为临床康复治疗方案制定等提供参考和帮助。
[Abstract]:The neuromuscular activity of the human body is accompanied by the release of corresponding electrophysiological signals.In recent years, the cooperative analysis of EEG and EMG has become a research hotspot in sports medicine, rehabilitation engineering and so on.Typical neuromuscular activity paradigms include autonomous action modes, stimulative action modes and imaginary action modes, and so on. Because of the different control mechanisms of the brain muscles, the evoked EEG and EMG signals also have different modes.By comparing and analyzing the relationship between electroencephalogram (EEG) and electromyography (EMG) in multiple motion modes, the relationship between motion and brain can be further understood.Taking flexion as an example, this paper designs three modes of action: autonomous, median nerve stimulation and imagination.According to the time interval of the action, two kinds of experimental schemes were set: long (10 s) and short (2 s). The scalp electroencephalogram (EEG) and the surface electromyography (EMG) of the flexor digitorum were simultaneously collected in the lead of C3/C4.After pre-processing of two-point peak threshold detection, the independent changes of EMG and EEG in different modes were studied using time-frequency analysis method based on Morlet wavelet.In this paper, the coherence of EEG signals under different action modes and time intervals is compared and analyzed by using biased directional coherence algorithm and improved correlation number extremum point analysis.The results show that the contralateral EEG dominates the EMG signal and interacts with the ipsilateral EEG in the autonomic action mode, while the contralateral EEG interacts with the ipsilateral EEG in the stimulation mode.EMG is the common information source of ipsilateral and contralateral EEG, and the effect on contralateral EEG is higher than that on ipsilateral EEG. In the imaginary action mode, EEG and EMG have no obvious causality, but there is information interaction between ipsilateral and contralateral EEG.In the same mode of action, the skew coherence of short interval is significantly higher than that of long interval, and the causality of stimulus is significantly higher than that of autonomous action, and the imagination action is the lowest at the same time interval.The phenomenon of 尾 -segment electromyoelectric coherence extremum concentration occurred in both short and long intervals, and the number of short interval coherence extremum was significantly higher than that of long interval.The results of time-frequency characteristic analysis also show that there are some differences between short and long time intervals, which preliminarily reveal the influence of time interval on coherence.Through the joint analysis of EEG signals in different time intervals and different modes of action, it is found that there is a high causality of EEG in short and stimulating modes, which may overlap with EEG signals in short motion.Brain neurons are more active and have strong dominance over limb movements in stimulation mode.The research results are helpful to better understand the motor ability and coordination ability of human neuromuscular system, and it is expected to provide reference and help for clinical rehabilitation treatment in the future.
【学位授予单位】：天津大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：R318.0

【相似文献】