CASIA OpenIR  > 毕业生  > 博士学位论文
基于扩散张量图像的脑白质研究
其他题名White Matter Analysis Based on Diffusion Tensor Image
龚高浪
学位类型工学博士
导师蒋田仔
2006-05-28
学位授予单位中国科学院研究生院
学位授予地点中国科学院自动化研究所
学位专业模式识别与智能系统
关键词扩散张量图像 白质纤维 扣带束 胼胝体 精神疾病 Diffusion Tensor Image White Matter Fasciculi Cingulum Corpus Callosum Psychiatric Diseases
摘要扩散张量成像(Diffusion Tensor Imaging)是一种新型的MR成像方式,它的发展使得对大脑白质纤维的无创伤性研究成为可能。基于扩散张量图像的脑白质研究已是现阶段脑影像研究领域的热门方向。本文的研究工作包括以下三个主要方面:针对扣带束的扩散属性研究方法及其应用、基于纤维形态的胼胝体压部的子分区方法以及扩散张量图像在部分精神疾病研究中的应用。 针对扣带束这一边缘系统的最大纤维,我们提出了一种基于弧度的、尺度不变的扣带束纤维参数化方法,从而能更准确的建立不同个体扣带束之间的解剖对应关系。基于该参数化方法,我们在正常右利手人群中沿着扣带束分析了左右侧扣带的FA属性值。统计结果发现:在正常右利手人群中存在明显左大于右的扩散异向性的不对称性,并且这种差异集中在除了最后端之外的扣带束中部的大部分区域。在此基础上,我们又征集了部分左利手的正常人,以考察利手因素对扣带束扩散属性的影响。在同样的研究方法框架下,统计结果表明:左利手的双侧扣带的扩散异向性都较右利手显著性偏低;并且在左利手人群中,依然存在显著性左大于右的扩散异向性不对称。 针对二维胼胝体结构的子分区问题,我们把胼胝体纤维的几何特性作为每个胼胝体体素的特征,从而实现对胼胝体压部的进一步子分区。这种基于纤维几何特征的胼胝体压部子分割方法是一种全新的子分割框架,初步的试验也取得了一些不错的结果。 通过与外单位精神科医师以及放射科医师的合作,我们将扩散张量图像应用到部分精神疾病患者的脑白质研究中,取得了一些很有意义的研究结果。1)通过对胼胝体的扩散张量图像分析,我们并未发现精神分裂症患者各个子分区的扩散属性值(即FA和ADC)与正常被试存在显著性差异,但是统计结果表明,胼胝体峡部的面积较正常被试显著性偏低;2)通过对部分首发精神分裂症患者与正常对照组全脑的白质扩散属性分析,我们发现精神分裂症患者在脑干、匡额区、额中回、颞下回、胼胝体压部、前扣带区域以及岛叶区域的白质FA值明显异常;3)通过对Alzheimer氏病患者与正常对照组的灰质密度和FA值的全脑关联分析,我们发现Alzheimer氏病患者的部分灰质萎缩区域与白质异向性下降区域在解剖位置上很相邻,但是也存在着一些在解剖上找不到相邻灰质萎缩区域的脑白质异常区域(主要是双侧上纵束、右侧顶上小叶、脑干),这可能提示了这些白质异常区域的变性是个相对独立的过程。
其他摘要This paper mainly includes three parts: Method development and its application for the analysis of cingulum; Subdivision of splenium using the traced fiber from tractography; Clinical application in some psychiatric diseases with diffusion tensor image. A scale-invariant parameterization method by arc-angle was proposed to study the cingulum, specifically, which is one of the most prominent white matter fiber tracts of the limbic system. This method can establish better anatomical correspondence of cingulum across subjects than the methods ever used. Using this improved fiber-based scheme, we found a significant left-greater-than-right asymmetry pattern in most segments of cingulum bundle, except the most posterior portion of cingulum, in DTI datasets of normal right-handed subjects. Further, another normal left-handed subjects were recruited to explore the microstructure of bilateral cingulum bundles associated with handedness. The statistical results showed a remarkable left-greater-than-right asymmetry pattern of anisotropy in most segments of cingulum bundles except the most posterior segment, for both right-handers and left-handers. Very interestingly, higher anisotropy of the right-hander than the left-hander was found in the bilateral cingulum bundles. An indirect approach was developed to subdivide the splenium of corpus callosum on mid-sagittal slice of diffusion tensor image (DTI). In this framework, the subdivision was implemented by clustering splenium voxels with some feature derived from the geometric properties of its crossing fiber. Some preliminary and reasonable result indicated the potential applicability of this approach. We have applied diffusion tensor image to the analysis of white matter in some psychiatric diseases, and some meaningful finding was obtained. 1). For the analysis of callosum corpus in schizophrenia, there were no significant differences between first-episode schizophrenics and controls in FA and ADC for each sub-regions of callosum corpus. But the isthmus area of the patients was significantly smaller than that of the controls. 2). For the analysis of whole brain in schizophrenia, we found that FA in white matter of the patients was lower than controls at the cerebral peduncle, frontal regions, inferior temporal gyrus, medial parietal lobes, hippocampal gyrus, insula, right anterior cingulum bundle and right corona radiata. 3). For the analysis of white matter and gray matter in Alzheimer’s disease, the results showed that the location of lower FA in AD patients were close to those of cortical volume reduction in some sites, but some white matter abnormalities (mainly the right superior parietal lobule, bilateral superior longitudinal fasciculi and cerebral peduncles) were structurally distant from the atrophic cortices. This indicated that white matter abnormality might be an independent factor in the progress of Alzheimer’s disease.
馆藏号XWLW985
其他标识符200418014690008
语种中文
文献类型学位论文
条目标识符http://ir.ia.ac.cn/handle/173211/5913
专题毕业生_博士学位论文
推荐引用方式
GB/T 7714
龚高浪. 基于扩散张量图像的脑白质研究[D]. 中国科学院自动化研究所. 中国科学院研究生院,2006.
条目包含的文件
文件名称/大小 文献类型 版本类型 开放类型 使用许可
CASIA_20041801469000(2100KB) 暂不开放CC BY-NC-SA请求全文
个性服务
推荐该条目
保存到收藏夹
查看访问统计
导出为Endnote文件
谷歌学术
谷歌学术中相似的文章
[龚高浪]的文章
百度学术
百度学术中相似的文章
[龚高浪]的文章
必应学术
必应学术中相似的文章
[龚高浪]的文章
相关权益政策
暂无数据
收藏/分享
所有评论 (0)
暂无评论
 

除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。