Non-invasive Multi-modal Parcellation of the Human Cerebral Cortex

描述（由申请人提供）：皮层区域构成人类皮层功能和解剖组织的重要和可接近的水平。在非人类灵长类动物的侵入性研究中，皮层区域被定义为功能差异（响应外部或内部刺激的神经活动模式）、结构（皮层每个区域内细胞的组织和连接）、连通性（不同区域之间连接的差异）和地形（某些皮质区域中存在的空间、身体等地图）。基于我们对猕猴的广泛了解和对人类的有限了解，人类大脑皮层包含150到200个皮质区域，可以使用多模态神经成像方法最好地识别。该项目的目的有两个：1)改进基于非侵入性核磁共振成像的结构技术，髓鞘映射，我们最近展示了可以在群体平均数据中定义许多皮质区域。工件破坏了组平均数据的某些区域，并且个人具有更多的工件。我们将使用改进的算法在更高质量的数据中检测和纠正这些伪影。预期的结果是更好的髓磷脂图谱和定义个体皮质区域的能力。2)在静息状态fMRI （R-fMRI）数据中识别可复制的皮质区域和区域边界。我们将改进基于梯度的R-fMRI分割，并将其应用于人类连接组项目获得的高分辨率7T R-fMRI数据。这个目标应该产生更详细的皮层区域和区域边界图，显示出跨扫描、受试者内部、跨受试者以及与其他模式相比的一致性。预期的结果是通过R-fMRI功能连接定义的更详细的皮质区域图，并与已发表的解剖和功能研究相关。将解剖学和功能数据定位到皮质区域的改进方法将大大有利于健康和疾病中大脑连接和功能的研究。准确定义个体和群体的皮质区域将使研究精神障碍的科学家能够更好地比较个体、群体和研究的结果。它还将使对那些与该疾病直接相关的皮质区域的具体研究成为可能。NIMH战略计划反复提到神经成像和人类大脑区域。通过改进人类皮层区域图谱，以及在非侵入性研究中准确定义个体区域的能力，将极大地帮助解释神经成像模式。皮质区域的准确定义也允许它们作为疾病的生物标志物（或作为测量其他非侵入性生物标志物的神经生物学基础标志，如髓磷脂含量、功能连通性或扩散束状图）。NIMH基础脑和行为科学优先项目也强调在神经成像研究中使用良好的解剖学信息，这也是本提案旨在提供的。

DESCRIPTION (provided by applicant): Cortical areas constitute an important and accessible level of functional and anatomical organization of the human cortex. In invasive studies of nonhuman primates, cortical areas have been defined using differences in Function (neural activity patterns in response to external or internal stimuli), Architecture (the organization of clls and connections within each area of the cortex), Connectivity (differences in connections between different areas), and Topography (maps of space, the body, etc present in some cortical areas). Based on our extensive knowledge in the macaque and limited knowledge in the human, the human cerebral cortex contains between 150 and 200 cortical areas that can best be identified using a mutli-modal neuroimaging approach. The project aims are twofold: 1) To improve upon a non-invasive MRI-based architectonic technique, myelin mapping, that we recently showed can define many cortical areas in group-average data. Artifacts corrupt some regions of the group average data and individuals have more artifacts. We will use improved algorithms to detect and correct these artifacts in higher quality data. The anticipated outcomes are better myelin maps and the ability to define cortical areas in individuals. 2) To identify reproducible cortical areas and areal boundaries in resting state fMRI (R-fMRI) data. We will improve gradient-based R-fMRI parcellation and apply it to high-resolution 7T R-fMRI data acquired by the Human Connectome Project. This aim should yield more detailed maps of cortical areas and areal borders that show consistency across scans, within subjects, across subjects, and when compared to other modalities. The anticipated outcome is a more detailed map of cortical areas defined by R-fMRI functional connectivity and related to published anatomical and functional studies. The study of brain connectivity and function in heath and disease will benefit greatly from improved methods for localizing anatomical and functional data to the cortical areas. Accurately defining cortical areas in individuals and groups will allow scientists studying mental disorders to better compare their results across individuals, groups, and studies. It will also enable specific study of those cortical areas directly implicated in the disease. The NIMH Strategic Plan mentions neuroimaging and human brain areas repeatedly. Interpreting neuroimaging patterns will be greatly aided by improved maps of human cortical areas across populations and the ability to accurately define areas in the individuals being studied non-invasively. Accurate definition of cortical areas also allows for their use as biomarkers of diseases (or as neurobiologically grounded landmarks for measuring other non-invasive biomarkers such as myelin content, functional connectivity, or diffusion tractography). The NIMH Priorities for Basic Brain and Behavioral Science also emphasizes the use of good anatomical information in neuroimaging studies, which this proposal aims to provide. PUBLIC HEALTH RELEVANCE: To understand the cerebral cortex in health and disease at a systems level, we need to know the identities of the component units, the cortical areas, and have a way of localizing experimental results to them. This project aims to provide advances towards both these goals. These advances will make it easier for investigators studying the brain in disease states to interpret their results and relate them to what is known about the cortical areas of interest.