ANALYSIS OF LARGE-SCALE BRAIN NETWORK IN HUMANS

这个子项目是许多研究子项目中利用 资源由NIH/NCRR资助的中心拨款提供。子项目和 调查员(PI)可能从NIH的另一个来源获得了主要资金， 并因此可以在其他清晰的条目中表示。列出的机构是 该中心不一定是调查人员的机构。 用于分析人类功能神经成像数据(例如，fMRI)的传统方法集中于孤立的大脑区域(例如，通过使用通用线性模型)或一小组大脑区域(例如，通过使用结构方程建模)。虽然这些方法提供了丰富的信息，但它们不能告诉我们整个大脑区域的交互性质。在这项初步研究中，我们的目标是通过使用有助于这种分析的方法(例如，结合图论使用的方法)来检查大规模的人脑网络。通常，在实验的一个时间点(一个典型的实验由数百个时间点组成)从一个受试者收集30,000个区域(以体素的形式)。在我们目前的设施中，计算要求很高，这使得分析非常困难。我们寻求利用网络基础设施伙伴关系提供的超级计算资源，对我们使用NIH基金收集的数据进行如此大规模的分析。

This subproject is one of many research subprojects utilizing the resources provided by a Center grant funded by NIH/NCRR. The subproject and investigator (PI) may have received primary funding from another NIH source, and thus could be represented in other CRISP entries. The institution listed is for the Center, which is not necessarily the institution for the investigator. Traditional methods for the analysis of human functional neuroimaging data (e.g., fMRI) focus on isolated brain regions (e.g., by using General Linear Model) or a small set of brain regions (e.g., by using Structural Equation Modeling). While informative, these methods fall short of informing us about the interactive nature of regions across the entire brain. In this pilot study, we aim at examining large-scale human brain networks by using methods that are conducive to such analysis (e.g., methods that are used in conjunction with graph theory). Typically, 30,000 regions (in the form of voxels) are collected from one subject at one time point of the experiment (a typical experiment consists of hundreds of time points). The high computational demands make the analysis extremely difficult in our current facility. We seek to use the super-computing resources provided by the CyberInfrastructure Partnership to perform such large-scale analyses on our data collected using NIH funds.