基于大鼠的SSVEP网络机制研究

发布时间：2018-01-18 03:06

本文关键词：基于大鼠的SSVEP网络机制研究　出处：《电子科技大学》2014年硕士论文　论文类型：学位论文

【摘要】：稳态视觉诱发电位(Steady state visual evoked potential,SSVEP)具有较高信噪比和稳定的频谱特征等优点,目前得到了广泛地应用。虽然已有的研究揭示SSVEP的产生涉及到多个脑区,但是其神经机制目前来说还不是很清楚,制约了相关领域的研究。在本文的研究中,采用了多通道,且兼有高时间、高空间分辨率的颅内记录方式,以此来记录麻醉状态下的大鼠脑电数据(electroencephalograph,EEG),从网络的层面上研究SSVEP的神经机制。本论文的主要工作如下:1.采用基于图论的脑网络分析方法,针对不同的刺激频率,我们研究了SSVEP的振幅与网络拓扑属性之间的关系。对于单只大鼠,具体分析了随时间动态变化的SSVEP的振幅与其对应的网络拓扑属性之间的关系,而在组间,具体分析了其空闲的基线网络属性与SSVEP响应的网络属性之间的关系,以及与SSVEP响应强度的关系。我们的结果首次揭示了SSVEP的产生与节点分布于全脑的网络重组有着密切的关系。在SSVEP的产生中,控制态的网络重组扮演着重要的角色,它可以为SSVEP的产生提供一个重要的预测指标。并且在与其他状态下的网络对比中发现,能够诱发出较强SSVEP的8Hz刺激下的网络拓扑结构连接更加紧密,并且枕叶与额叶之间的长程连接明显增多,所以我们推测,枕叶和额叶之间的信息交互在SSVEP的生成中扮演着重要角色。2.首次利用基于神经参数集总模型耦合构成的双柱模型,对在SSVEP响应中的额叶和枕叶间的信息交互进行了研究,揭示了不同频率刺激下SSVEP响应的机制,并且利用偏有向相干分析方法(Partial directed coherence,PDC),对真实的EEG数据作有向网络分析,对双柱模型的结果进行验证。在本文的研究中,我们采用粒子群方法(Particle Swarm Optimization,PSO)对不同刺激状态下的模型参数进行估计,采用与真实脑电数据误差最小的输出参数,然后基于该参数来对相应SSVEP的机制进行解释。结果显示,对于不同的频率刺激,在枕叶和额叶内部的连接没有显著变化,并且额叶向枕叶的信息反馈也没有明显的变化,最主要的变化来自于枕叶向额叶的信息传递,也就是说,在有SSVEP响应的8Hz刺激条件下,显示出枕叶较强地向额叶信息的传递。这一发现揭示枕叶向额叶的信息传递强度对SSVEP的产生有较大的影响,而且这与利用PDC构建有向网络得到的结果是一致的。
[Abstract]:Steady state visual evoked potential. SSVEP has the advantages of high signal-to-noise ratio (SNR) and stable spectral characteristics, and has been widely used, although some studies have revealed that the production of SSVEP involves multiple brain regions. However, the neural mechanism is not very clear, which restricts the research in related fields. In this study, we use multi-channel, and have high time, high spatial resolution intracranial recording. The electroencephalographic data of anesthetized rats were recorded. The main work of this thesis is as follows: 1. The brain network analysis method based on graph theory is used to study the neural mechanism of SSVEP. We studied the relationship between the amplitude of SSVEP and the topological properties of the network. The relationship between the amplitude of SSVEP and its corresponding network topology attributes is analyzed in detail. The relationship between its idle baseline network attribute and the network attribute of SSVEP response is analyzed in detail. Our results show for the first time that the generation of SSVEP is closely related to the network recombination of nodes distributed throughout the brain. The network recombination of control state plays an important role, it can provide an important predictor for the generation of SSVEP, and it is found in the comparison with other states of the network. It can induce a stronger SSVEP 8Hz stimulation of the network topology connections more closely, and the occipital lobe and the frontal lobe between the long distance connections significantly increased, so we speculate. The information interaction between occipital lobe and frontal lobe plays an important role in the generation of SSVEP. 2. A two-column model based on neural parameter lumped model is used for the first time. The information exchange between frontal lobe and occipital lobe in SSVEP response was studied, and the mechanism of SSVEP response under different frequency was revealed. And the partial directed coherence analysis method is used to analyze the real EEG data in directed network. The results of the two-column model are verified. In this study, we use particle Swarm Optimization. PSOs are used to estimate the model parameters under different stimulus states, and the output parameters with the smallest error from the real EEG data are used to explain the mechanism of the corresponding SSVEP based on these parameters. The results show that. For different frequency stimuli, the connection between the occipital lobe and the frontal lobe did not change significantly, and the information feedback from the frontal lobe to the occipital lobe did not change significantly, and the most important change came from the information transmission from the occipital lobe to the frontal lobe. That is to say, under the condition of 8Hz stimulation with SSVEP response. The findings show that the occipital lobe strongly transmits information to the frontal lobe. This finding reveals that the intensity of information transmission from occipital lobe to frontal lobe has a great influence on the production of SSVEP. And this is consistent with the result of using PDC to build a directed network.
【学位授予单位】：电子科技大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：R338

【相似文献】