静息态精神分裂症脑网络及任务态语义脑磁信号能量分布研究

发布时间：2018-04-24 14:29

本文选题：脑磁信号 + 精神分裂症　；参考：《南京邮电大学》2017年硕士论文

【摘要】：大脑有多个功能区,具有复杂的处理机制,脑区之间相互作用,实现语言、行为、思维等功能。脑磁信号是大脑产生的一种磁场,具有高空间分辨率等脑电不具有的特点,包含了丰富的生理信息。静息态和任务态下的脑磁信号分别表征了大脑的不同活动状态,是脑研究的两个主要类别。精神分裂症是一种常见的精神疾病,改变了病人大脑的作用机理,研究发现,静息态能够反映大脑的本征状态,有利于发现患者的功能异常。对于患者来说,进行任务有时较为困难,故研究了正常人的任务态,针对特定的任务——语义研究,分析不同语义条件下的脑磁信号,找出大脑处理特定任务具有显著差异的主要区域。论文分别研究了静息态精神分裂症患者MEG和正常人不同语义条件下的MEG,基于复杂网络理论,研究了两种脑网络构建方法。在精神分裂症患者静息态脑磁数据的基础上,分别分析脑区域间的相关性及颞叶区内通道间相关性,构建MEG相关网络,并分析网络特征参数,探究患者与正常人网络特征的差异;在语言语义数据上,研究能量在头部分布特点以及变化过程,找出具有显著差异的通道,分析大脑对不同语义做出的相应处理。主要工作如下:(1)根据大脑区域分布,将脑功能区作为节点,提出用可以度量非线性差异的斯皮尔曼秩次相关系数来计算区域节点间的相关性,构建MEG功能网络,分析网络特征参数,精神分裂症患者某些区域可能受到了损伤。(2)以往研究多是构建整个脑网络,文章针对具有显著差异的颞叶区,计算格兰杰因果关系,创新性地构建MEG有向二值网络,计算脑网络特征,找出网络的关键节点,比较精神分裂症患者和正常人的网络差异。(3)研究不同语义条件下的能量分布,以及能量分布的变化过程,针对差异较大的初始不一致和初始相等条件,计算平面梯度,进行统计检验,找出具有显著差异的通道。分析处理不同语义的句子大脑皮质的差异。
[Abstract]:The brain has multiple functional regions, with complex processing mechanisms, interaction between brain regions, the realization of language, behavior, thinking and other functions. Brain magnetic signal is a kind of magnetic field produced by the brain. It has the characteristics of high spatial resolution and other EEG signals, and contains abundant physiological information. The brain magnetic signals in resting state and task state represent different brain activity respectively, and are two main types of brain research. Schizophrenia is a common mental disease, which changes the mechanism of brain function of patients. It is found that resting state can reflect the intrinsic state of brain and is helpful to discover the abnormal function of patients. For patients, the task is sometimes difficult, so the task state of the normal person is studied. According to the specific task-semantic research, the brain magnetic signals under different semantic conditions are analyzed. Identify major areas in which the brain processes specific tasks with significant differences. In this paper, we studied the MEG of patients with resting schizophrenia and the normal subjects under different semantic conditions. Based on the complex network theory, we studied two kinds of brain network construction methods. Based on the resting magnetic data of schizophrenic patients, the correlation between brain regions and temporal lobe channels was analyzed, and the MEG correlation network was constructed, and the characteristic parameters of the network were analyzed. In terms of language and semantic data, the distribution of energy in the head and the process of change are studied to find out the channels with significant differences, and to analyze the corresponding processing of different semantics in the brain. The main work is as follows: (1) according to the distribution of the brain region, the functional area of the brain is regarded as the node, and the Spelman rank correlation coefficient, which can measure the nonlinear difference, is used to calculate the correlation between the regional nodes, and the functional network of MEG is constructed. According to the characteristic parameters of the network, some areas of schizophrenia may be injured. (2) previous studies were mostly about the construction of the whole brain network. Granger causality was calculated for the temporal lobe with significant differences. We creatively construct MEG directed binary network, calculate the characteristics of brain network, find out the key nodes of the network, compare the network difference between schizophrenic patients and normal people, and study the energy distribution under different semantic conditions. According to the initial inconsistency and the initial equality condition, the plane gradient is calculated, and the statistical test is carried out to find out the channel with significant difference. Analyze the differences in the cerebral cortex of sentences with different semantics.
【学位授予单位】：南京邮电大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：R749.3;O157.5

【参考文献】