神经元锋电位与场电位之间锁相性分析的新方法及其应用

发布时间：2018-10-29 20:05

【摘要】：神经元动作电位(锋电位)的发放与局部场电位(local field potentials,LFP)节律之间的锁相关系反映了神经元参与大脑某种活动的信息,具有神经编码的特性。研究这种锁相关系时,目前常用的方法是区间统计法(简称Bin法)。它可以定性判断锁相性是否存在,但不能定量区别各种锁相性的强弱。此外,Bin法还存在多种缺陷,例如需要将场电位人为分割成窄频带信号,这有可能破坏场电位原本包含的重要信息；计算效率较低,在分析多种不同节律时需要重复计算；以相位分区的方式统计落在每个区间上的锋电位个数时,设定的相位区间大小也会直接影响计算结果。为了弥补Bin法的不足之处,本文利用锋电位触发的叠加平均信号(spike-triggered average,STA)提出了一种判断锁相性的新方法。将原始信号经叠加平均计算得到的信号(即STA波)与原始场电位信号的功率百分比作为定量指标,来评价锋电位与场电位节律之间锁相性的强弱。本文分析了麻醉大鼠海马CA1区的大量实验数据,发现该功率百分比的值与锁相性强弱之间存在单调关系。为了进一步考察该指标与锁相性强弱的关系,本文利用仿真数据研究,结果表明,该指标能够提供有效的界值来区分锋电位与某场电位节律是否锁相,并且它的值越大,锁相关系就越强。在确定锁相的神经元后,还可利用STA波直接计算锁相神经元锋电位在场电位上发放的相位平均角。本文利用16通道微电极阵列记录海马CA1区不同分层上的场电位,应用STA法分析了CA1区神经元锋电位发放与顶树突层、胞体层和基树突层上场电位的θ节律和γ节律之间的关系,发现锋电位在不同神经组织结构分层的不同场电位节律上发放的相位角是不同的。在计算锋电位发放相位平均角时,本文提出的STA法与Bin法相比具有显著的高效性。STA波经任意所需频段的单次滤波后就可以得到锋电位在场电位某特定节律上发放的相位平均角,无需重复进行叠加平均计算。而Bin法却需要分别统计每个锋电位在所需的场电位某特定节律波上发放的相位分布,再经圆周统计法计算相位平均角。因此,STA法具有较高的计算效率。综上所述,本文提出了分析神经元锋电位发放与场电位节律之间关系的STA法,并利用STA波设计了计算锋电位与场电位节律锁相性强弱的定量参数,避免了Bin法中事先滤波和人为设定场电位频率对分析结果的影响。这种STA法对于大脑神经信息编码机制的研究具有重要意义。
[Abstract]:The phase-locked relationship between the release of action potential (spike) and the rhythm of local field potential (local field potentials,LFP) reflects the information that neurons participate in some activities of the brain and possess the characteristics of neural coding. At present, interval statistics (Bin method) is commonly used to study the phase locked relation. It can qualitatively judge whether the phase locking exists or not, but it can not quantitatively distinguish the strength of various phase-locking properties. In addition, the Bin method also has many defects, such as the need to artificially divide the field potential into narrow band signals, which may destroy the important information originally contained in the field potential, and the calculation efficiency is low, so it is necessary to repeat calculation when analyzing many different rhythms. When the number of spikes falling on each interval is counted by phase partitioning, the size of the set phase range will also directly affect the calculation results. In order to make up for the deficiency of Bin method, a new method to judge the phase locking property is proposed by using the superposition average signal (spike-triggered average,STA) triggered by the spike potential. The power percentage of the original signal (i.e. STA wave) and the original field potential signal is used as a quantitative index to evaluate the phase locking between the spike potential and the field potential rhythm. A large number of experimental data of the hippocampal CA1 region in anesthetized rats were analyzed. It was found that there was a monotone relationship between the power percentage and the strength of phase locking. In order to further investigate the relationship between this index and the strength of phase locking, the simulation data are used in this paper. The results show that the index can provide effective bounds to distinguish whether the rhythm of spike potential and a field potential is phase-locked or not, and the larger the value is, the greater the value is. The stronger the phase locking relationship is. After determining the lock-in neurons, the phase mean angle on the field potential of the lock-in neuron can be directly calculated by using the STA wave. In this paper, the field potentials on different layers of hippocampal CA1 region were recorded by a 16-channel microelectrode array. The relationship between the spike discharge of neurons in the CA1 area and the 胃 rhythm and 纬 rhythm of the parietal dendritic layer, the somatic layer and the basal dendritic layer was analyzed by STA method. It is found that the phase angle of spike distribution is different in different field potential rhythms of different neural tissue structures. In calculating the mean angle of the phase of the spike, The STA method proposed in this paper is more efficient than the Bin method. After a single filtering of the STA wave in any required frequency band, the phase mean angle on a specific rhythm of the potential field potential can be obtained, and the superposition average calculation is not necessary. However, the Bin method needs to calculate the phase distribution of each spike on a specific rhythmic wave of the required field potential, and then calculate the phase mean angle by circumferential statistics. Therefore, STA method has high computational efficiency. To sum up, a STA method is proposed to analyze the relationship between spike potential release and field potential rhythm, and the quantitative parameters are designed by using STA wave to calculate the strong and weak phase locking characteristics of the spike potential and field potential rhythm. The effect of pre-filtering and artificially setting field potential frequency on the analysis results is avoided in Bin method. This STA method is of great significance for the study of neural information coding mechanism.
【学位授予单位】：浙江大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：TN911.7;R338.1

【参考文献】