Metabolic connectivity by interregional correlation analysis using statistical parametric mapping (SPM) and FDG brain PET; methodological development and patterns of metabolic connectivity in adults
Metabolic connectivity by interregional correlation analysis using statistical parametric mapping (SPM) and FDG brain PET; methodological development and patterns of metabolic connectivity in adults
复制标题
DOI：
10.1007/s00259-008-0808-z
复制
发表时间：
2008-09-01
影响因子：
9.1
通讯作者：
Lee, Myung Chul
中科院分区：
文献类型：
--
作者：
Lee, Dong Soo;Kang, Hyejin;Lee, Myung Chul
目的用氟脱氧葡萄糖正电子发射断层扫描（FDG-PET）检测静息脑的区域连接区。研究了体素代谢连通性，并通过对FDG-PET数据的统计参数映射进行区域间相关性分析，建立了规范性数据。研究材料和方法种子感兴趣体积（VOIs）作为功能脑单元的特征，包括它们的位置、大小和独立的测定方法。种子脑区被确定为基于群体的回状VOIs （n = 70）或基于群体的细胞结构Brodmann区（BA, n = 28）。这些地区的FDG摄入量被用作一般线性模型中的自变量，以搜索与平均种子VOI计数相关的体素。在整个大脑区域寻找正相关。结果在正常成年人中，三分之一的回转体素产生了局限于自身的相关性，但在其他人中，相关体素扩展到邻近区域和/或对侧同源区域。在这些具有广泛连通性的区域中，在中线上发现了相关体素，并且在左右同源种子voi的连通性模式中观察到不对称。大多数可用的BAs产生的相关性达到对侧同源区域和/或邻近区域。代谢连通性的程度与种子VOI大小或用于定义种子VOI的方法无关。结论脑功能单元的代谢连通性模式依赖于它们的区域位置。我们提出FDG-PET数据的区域间相关性分析提供了一种检查静息人脑体素区域代谢连通性的方法。
Purpose Regionally connected areas of the resting brain can be detected by fluorodeoxyglucose-positron emission tomography (FDG-PET). Voxel-wise metabolic connectivity was examined, and normative data were established by performing interregional correlation analysis on statistical parametric mapping of FDG-PET data.Materials and methods Characteristics of seed volumes of interest (VOIs) as functional brain units were represented by their locations, sizes, and the independent methods of their determination. Seed brain areas were identified as population-based gyral VOIs (n = 70) or as population-based cytoarchitectonic Brodmann areas (BA; n = 28). FDG uptakes in these areas were used as independent variables in a general linear model to search for voxels correlated with average seed VOI counts. Positive correlations were searched in entire brain areas.Results In normal adults, one third of gyral VOIs yielded correlations that were confined to themselves, but in the others, correlated voxels extended to adjacent areas and/or contralateral homologous regions. In tens of these latter areas with extensive connectivity, correlated voxels were found across midline, and asymmetry was observed in the patterns of connectivity of left and right homologous seed VOIs. Most of the available BAs yielded correlations reaching contralateral homologous regions and/or neighboring areas. Extents of metabolic connectivity were not found to be related to seed VOI size or to the methods used to define seed VOIs.Conclusions These findings indicate that patterns of metabolic connectivity of functional brain units depend on their regional locations. We propose that interregional correlation analysis of FDG-PET data offers a means of examining voxel-wise regional metabolic connectivity of the resting human brain.