基于独立成分分析脑卒中运动功能障碍患者的多网络静息态功能连接异常研究

发布时间：2018-01-13 15:07

本文关键词：基于独立成分分析脑卒中运动功能障碍患者的多网络静息态功能连接异常研究　出处：《华东师范大学》2017年硕士论文　论文类型：学位论文

【摘要】：脑卒中后超过半数以上的病人都会遗留一定程度的运动功能障碍,对患者的日常生活十分不利。静息态功能磁共振成像(resting-state functional magnetic resonance imaging,rs-fMRI)因其简单易行的操作模式和对人脑自发神经活动的有效解读,现已成为研究脑卒中运动功能恢复神经机制的一种有效手段。越来越多的rs-fMRI功能连接研究表明,感觉运动网络是易遭受卒中病变攻击损伤的核心网络。与感觉运动功能相关的运动高级认知调控网络,例如额顶网络、执行控制网络及运动注意网络等也会发生功能损伤或代偿,即所谓功能重组。基于数据驱动的独立成分分析(independent component analysis,ICA)具有构建多个网络的功能,即选择网络适宜的成分数,可以同时考察脑卒中病变导致的这些多网络变化,甚至能够进一步弄清这些网络之间的相互作用影响,这即是本课题研究关注的重点。课题研究主要内容包括:(1)收集33名慢性期左侧皮层下脑卒中偏瘫患者和34名年龄、性别相匹配的健康被试的rs-fMRI数据,采用组ICA方法提取出健康对照组与脑卒中患者组的感觉运动网络并比较其差异。结果发现,ICA提取出的患者组的感觉运动网络(背侧、腹侧、左侧和右侧感觉运动网络)均显示功能连接明显下降,这说明卒中病变对偏瘫患者的感觉运动系统造成了多范围损伤。(2)同样以该组被试为研究对象,采用组ICA方法提取出11个静息态功能网络。与健康对照组比较,患者组出现多个网络功能连接异常,包括感觉运动网络、视觉网络、背侧注意网络、右额顶网络和执行控制网络。其中背侧注意网络内的顶下小叶功能连接指数与患手运动功能评分(fugl-meyer assessment,FMA)值呈正相关(r = 0.357,p=0.041),而执行控制网络中的顶下小叶的功能连接指数与患手运动功能评分FMA值呈负相关(r =-0.518,p = 0.002)。网络间分析进一步发现,背侧注意网络与感知觉网络(感觉运动网络、运动网络、视觉网络)的功能连接显著降低,默认网络与右额顶网络的功能连接显著升高,但默认网络后端与视觉网络的功能连接显著降低,而执行控制网络与右额顶网络的功能连接几乎消失。以上结果证实了脑卒中偏瘫患者的静息态多网络内和网络间功能连接异常,尤其是背侧注意网络的功能明显下降和执行控制网络的过度功能代偿对偏瘫患者手运动功能恢复并非好的征象。
[Abstract]:After stroke, more than half of the patients have left motor dysfunction to a certain extent, is very detrimental to the daily life of the patients. Resting state functional magnetic resonance imaging (resting-state functional magnetic resonance imaging, rs-fMRI) because of its simple operation mode and the spontaneous neural activity of the brain and effective interpretation, has become a kind of effective means of nerve study on the mechanism of recovery of motor function of stroke. More and more rs-fMRI functional connectivity studies show that the sensorimotor network is vulnerable to the core network of stroke attack damage. Sports advanced cognitive regulation network related sensorimotor function, such as the frontal parietal network, executive control network and the network will pay attention to injury or compensatory function that is, the so-called functional reorganization. Independent component analysis based on data driven (independent component analysis, ICA) with Construction of a plurality of network functions, namely network selection scores into appropriate, can these network changes while studying stroke lesions result, even can affect the interaction between the further understanding of these networks, which is the focus of this research focus. The main contents of this research include: (1) collection of hemiplegic stroke patients 33 chronic left cortex and 34 rs-fMRI data of age, gender matched healthy subjects, extract the sensorimotor network healthy control group and stroke group and compared group by ICA method. The results showed that patients with sensorimotor network ICA extract (dorsal, ventral side, left and right sensorimotor network) showed decreased functional connectivity, indicating that the lesions of hemiplegic patients with stroke sensorimotor system caused by multi range damage. (2) in the same group as the research subjects of As the group ICA, extracted 11 resting state functional network. Compared with healthy controls, patients with abnormal function of multiple network connections, including sensorimotor network, visual network, dorsal attention network, network and executive control network. The dorsal frontoparietal attention function in the network under the top lobule connecting the hand movement function index and risk score (Fugl-Meyer assessment, FMA) were positively correlated (r = 0.357, p=0.041), and the executive control network in the inferior parietal lobule function connecting the hand movement function index and the risk of FMA score was negatively correlated (R =-0.518, P = 0.002). Further analysis showed that the network. The dorsal attention network and perception network (sensorimotor network, sports network, visual network connectivity) decreased significantly, the default network and the right frontoparietal network connectivity increased significantly, but the default network terminal and the visual function of the network connection Decreased significantly, but the executive control network and the right frontoparietal network connectivity almost disappeared. The above results confirmed the abnormal resting state of stroke patients in the multi network and inter network connection function, especially the dorsal attention over compensatory function of network function decreased significantly and executive control network of hand motor function in patients with hemiplegia recovery is not a good sign.

【学位授予单位】：华东师范大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：R743.3;O157.5

【相似文献】