CASIA OpenIR  > 毕业生  > 硕士学位论文
融合结构和功能影像研究阿尔茨海默病脑网络异常
窦雪娇
Subtype硕士
Thesis Advisor刘勇研究员
2019-05-31
Degree Grantor中国科学院大学
Place of Conferral北京
Degree Discipline模式识别与智能系统
Keyword阿尔茨海默病,弥散张量成像,功能磁共振成像,白质纤维束,默认网络
Abstract

阿尔茨海默病(Alzheimer's disease,AD)是一种以渐进性痴呆为主要特征的慢性神经系统退行性疾病,而遗忘型轻度认知障碍(amnestic mild cognitive impairment,aMCI)被定义为正常老年人和AD患者之间的过渡状态。功能磁共振成像可以反映脑皮质对外界刺激响应的血氧功能,弥散张量成像已广泛用于结构完整性的研究并刻画AD患者的白质病变。默认网络(default mode network,DMN)参与许多脑功能和活动,其完整性在AD中得到了广泛的研究。本文主要研究AD患者的白质(white matter,WM)纤维束的异常和基于默认网络(default mode network,DMN)分析结构和功能连接的异常模式以及探索结构-功能的耦合关系。

为了分析AD/aMCI患者的白质纤维束的完整性,我们采用了自动纤维束量化(automated fiber quantification,AFQ)的方法。AFQ是一种可以快速且可靠地识别大脑内主要的白质纤维束,并量化其扩散特征的完全自动化的方法。本部分研究的主要目的是评估aMCI与AD患者以及健康对照组(normal controls,NCs)中白质的完整性和异常性。为此,我们首先通过AFQ方法来识别20条主要的白质纤维束,基于发现数据集中的120个被试(39个NCs,34个aMCI和47个AD)评估白质完整性和异常,并在重复数据集的122个被试(43个NCs,37个aMCI和42个AD)中进行了结果重复性的验证。在发现数据集中通过组间比较,识别沿着白质纤维束位置的逐点差异,同时在重复数据集中进行验证。然后,我们以沿着白质纤维束对应的扩散张量指标为特征,利用支持向量机对AD和NCs进行分类,并进行多次交叉验证。相关性分析结果显示,微结构白质的改变和分类器输出对应的伪概率与患者组的认知能力水平高度相关,这表明可信且稳定的早期生物标志物可能有助于AD的临床应用。本部分的研究为探索白质完整性以及AD的临床应用提供了很好的方法和流程,可能有助于其他神经和精神障碍疾病的研究。

对于分析AD患者默认网络内结构连接和功能连接的异常,我们基于发现数据集和重复数据集分别构造了相应的结构和功能矩阵。在默认网络中的几条结构连接对应的平均FA/MD值检测到显著的组间差异,而相比于NCs,AD患者在默认网络的28条连接中有16条发现了降低的功能连接。上述提及的大部分的异常连接在aMCI和AD组中均与认知能力评分存在一定的相关性。我们还发现aMCI与AD组在后扣带与右侧海马体之间的结构和功能连接之间是存在共变趋势的。我们以默认网络为整体分析了AD患者异常的结构和功能连接,有助于对AD/aMCI 的默认网络连接模式的理解。

Other Abstract

Alzheimer's disease (AD) is a chronic neurodegenerative disease characterized by progressive dementia, and amnestic mild cognitive impairment (aMCI) has been defined as a transitional state between normal aging and Alzheimer's disease. Functional magnetic resonance imaging (fMRI) can reflect the blood oxygen function of the cerebral cortex responding to external stimulus, and diffusion tensor imaging (DTI) has been widely used to show structural integrity and delineate white matter (WM) degeneration in AD. The default mode network (DMN) is known to be involved in many brain functions and activities and its integrity was widely investigated in AD. This thesis mainly studied the abnormalities of white matter tracts in AD patients, investigated abnormal structural and functional connectivity pattern based on default network and explored structural-functional coupling.

In order to analyze abnormal white matter tracts in AD/aMCI patients, we employed the automated fiber quantification (AFQ) method. AFQ is a fully automated method that can rapidly and reliably identify major WM fiber tracts and evaluate white matter properties. The main aim of this study was to assess WM integrity and abnormalities in a cohort of patients with aMCI and AD as well as normal controls (NCs). For this purpose, we first used AFQ to identify 20 major WM tracts and assessed WM integrity and abnormalities in a cohort of patients with 120 participants (39 NCs, 34 aMCI and 47 AD) in the discovery dataset, as well as 122 participants (43 NCs, 37aMCI and 42 AD) in the replicated dataset. Pointwise differences along the WM tracts were identified in the discovery dataset and simultaneously confirmed by the replicated dataset. Next, we investigated the utility of DTI measures along WM tracts as features to classify patients with AD from NCs with multilevel cross validations using a support vector machine. The correlation analysis revealed that the identified impaired microstructural WM alterations and classification output were highly associated with cognitive ability in the patient groups, indicating that this credible, robust potential early biomarker may be useful for clinical application in AD. This systematic study provides a pipeline to examine WM integrity and its potential clinical application in AD and may be useful in studying other neurological and psychiatric disorders.

For the analysis of estimating structural and functional connectivity abnormities in AD within DMN, we derived FC and SC matrices based on a cohort of patients in the discovery and replicated dataset, respectively. The significant group differences were detected for the mean FA/MD of several structural connections, while 16 out of 28 connections within DMN were detected lower functional connectivity in AD compared to NCs. And most of the abnormal connections mentioned above were associated with the cognitive ability scores in the aMCI and AD groups. We also found that structural-functional decoupling was existed between posterior cingulate cortex (PCC) and right hippocampus in aMCI and AD groups. We analyzed the abnormal structural and functional connectivity in AD within DMN, and contributed to the understanding of default mode network connectivity in AD/aMCI.

Pages100
Language中文
Document Type学位论文
Identifierhttp://ir.ia.ac.cn/handle/173211/23778
Collection毕业生_硕士学位论文
Recommended Citation
GB/T 7714
窦雪娇. 融合结构和功能影像研究阿尔茨海默病脑网络异常[D]. 北京. 中国科学院大学,2019.
Files in This Item:
File Name/Size DocType Version Access License
中国科学院自动化研究所硕士学位论文_窦雪(3991KB)学位论文 暂不开放CC BY-NC-SAApplication Full Text
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[窦雪娇]'s Articles
Baidu academic
Similar articles in Baidu academic
[窦雪娇]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[窦雪娇]'s Articles
Terms of Use
No data!
Social Bookmark/Share
All comments (0)
No comment.
 

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.