Assessing Large-scale Connectivities in Prenatal Cocaine Exposure Affected Brains

描述(申请人提供)：本申请是对PA-10-067(研究项目资助(家长R01))的响应。问题陈述：最近文献中的神经成像研究，包括我们自己的研究，表明在产前可卡因暴露(PCE)影响的大脑中，包括唤醒调节、工作记忆、语言、执行功能、注意力和视觉系统在内的几个大脑网络都发生了改变。然而，在受PCE影响的青少年中，大规模脑网络中结构和功能连接性的改变以及结构脑结构的改变在很大程度上是未知的。定量评估PCE青少年和对照组的大规模大脑连接性的主要障碍是严重缺乏密集的大脑地标，这些地标在不同的大脑中是一致的，并且可以作为连接图谱的公共网络节点。方法：最近，我们创建并验证了一种变革性的数据驱动方法，该方法在健康的年轻成年人大脑中发现了358个一致的大脑地标的密集地图，称为密集个性化和基于公共连接的皮质地标(DICCCOL)，每个地标由来自扩散张量成像(DTI)数据的一致白质纤维连接模式定义。我们的验证结果表明，这358个由DTI派生的DICCCOL地标在不同的数据集中具有惊人的重复性，具有有意义和准确的功能定位，并具有自动建立的跨对象对应关系。重要的是，使用DTI数据可以在新的单一大脑中准确有效地预测这358个大脑里程碑。因此，这组358个密集地标为大规模结构和功能连通性评估提供了通用网络节点。在这个项目中，我们建议将我们的DICCCOL模型及其预测框架应用于现有的Emory PCE/CONTROL数据集，并评估受PCE影响的青少年的大规模连接性。意义：1)发现的常见DICCCOL图谱，连同一致的结构连接模式，可以被认为是下一代脑图谱，它将比使用了100多年的Brodmann脑图谱具有更细微的粒度和更好的功能同质性。2)基于开源平台Insight Toolkit(ITK)和DICCCOL MAP的DICCCOL预测框架的传播将有助于在依赖于脑ROI精确定位的脑成像中的众多应用。3)PCE中DICCCOL标志物的大规模脑网络的连接性改变在很大程度上是未知的。在这个项目中，通过评估多模式DTI中的大规模网络和PCE/对照大脑的静止状态fMRI数据集，将显著弥合这一知识鸿沟。 公共卫生相关性：简而言之，该项目旨在将UGA密集个体化和基于共同连接性的皮质地标(DICCCOL)模型和里程碑预测框架应用于现有的Emory产前可卡因暴露(PCE)/对照数据集，并评估PCE青少年的大规模连接性。该项目旨在预测PCE/对照中的DICCCOL图谱，在三个种群中发现共同的DICCCOL图谱，评估PCE/对照中的大规模结构连接性，并评估PCE/对照中的大规模功能连接性。其结果将是对PCE对广泛的连接性改变的致畸作用的新阐明，以及一系列基于连接性的标志物的集合，这些标志物可以预测PCE的临床和行为测量。

DESCRIPTION (provided by applicant): This application is in response to the PA-10-067 (Research Project Grant (Parent R01)). Problem statement: Recent neuroimaging studies in the literature, including our own ones, have shown that several brain networks, including arousal regulation, working memory, language, executive function, attention, and vision systems are altered in prenatal cocaine exposure (PCE) affected brains. However, the alterations of structural and functional connectivities in large-scale brain networks and the alterations of structural brain architecture in PCE affected adolescents are largely unknown. The major barrier to the quantitative assessments of large-scale brain connectivities in PCE adolescents and controls is the critical lack of dense brain landmarks that are consistent across different brains and can serve as common network nodes for connectivity mapping. Approaches: Recently, we created and validated a transformative data-driven approach that discovered a dense map of 358 consistent brain landmarks, called Dense Individualized and Common Connectivity-based Cortical Landmarks (DICCCOL), in healthy young adult brains, each of which is defined by consistent white matter fiber connectivity pattern derived from diffusion tensor imaging (DTI) data. Our validation results have shown that these 358 DTI-derived DICCCOL landmarks are strikingly reproducible in separate datasets, have meaningful and accurate functional localizations, and possess automatically-established cross-subjects correspondences. Importantly, these 358 brain landmarks can be accurately and efficiently predicted in a new, single brain with DTI data. Therefore, this set of 358 dense landmarks offers common network nodes for large-scale structural and functional connectivities assessments. In this project, we propose to apply our DICCCOL models and their prediction framework on existing Emory PCE/control datasets and assess large-scale connectivities in PCE affected adolescents. Significance: 1) The discovered common DICCCOL map, together with the consistent structural connection patterns, can be considered and used as the next-generation brain atlas, which will have much finer granularity and better functional homogeneity than the Brodmann brain atlas that has been used for over 100 years. 2) The dissemination of the DICCCOL prediction framework based on the open source platform of Insight Toolkit (ITK) and the DICCCOL map will contribute to numerous applications in brain imaging that rely on accurate localization of brain ROIs. 3) Connectivity alterations in large-scale brain networks of DICCCOL landmarks in PCE are largely unknown. This knowledge gap will be significantly bridged in this project by assessing large-scale networks in multimodal DTI and resting state fMRI datasets of PCE/control brains. PUBLIC HEALTH RELEVANCE: Briefly, this project aims to apply UGA Dense Individualized and Common Connectivity-based Cortical Landmarks (DICCCOL) models and landmark prediction framework on existing Emory prenatal cocaine exposure (PCE)/control datasets and assess large-scale connectivities in PCE adolescents. This project aims to predict DICCCOL maps in PCE/controls, discover common DICCCOL maps in three populations and assess large-scale structural connectivities in PCE/controls, and assess large-scale functional connectivities in PCE/controls. The outcome will be novel elucidations of PCE's teratogenic effects on widespread connectivity alterations and a collection of connectivity-based markers that are predictive of PCE clinical and behavior measurements.