Imaging platform and computational HARDI atlas of the human hippocampus

摘要 海马体结构是神经影像学中的一个重要结构，因为它在广泛的 一系列神经疾病，包括癫痫、精神分裂症和阿尔茨海默病(AD)。三个- 对复杂的海马区内组织和子区连通性的维度映射至关重要 重要，因为已知海马体的不同亚区表现出选择性和差异性 易患不同的神经系统疾病，如癫痫和阿尔茨海默病。尽管它在主要的 神经紊乱、人类海马层内微结构和亚场连通性 仍然很难用三维成像技术来描绘。 R01计划的目标是开发第一个三维弥散磁共振成像计算图谱 完整的人类海马体。拟议的地图集将建立在关键技术发展的基础上，使 11.7T的完整固定海马区的高空间-角分辨率扩散磁共振成像(DMRI)。这个 基于多个海马区的高角分辨率扩散成像(HARDI)数据生成的图谱 标本将史无前例地描绘人类海马区内的组织和回路 微观结构细节。此外，目前对海马体回路及其在疾病中的破坏的理解是 主要基于二维组织切片。在此建立的影像平台和dMRI图谱 将首次实现对区域内特定变化的调查和三维绘图 正常衰老和阿尔茨海默病脑的海马区结构，提供了海马区病理的细节 比以前可能的更详细。我们将通过四个具体目标实现这些目标：(一)加快发展 用于组合(k，q)空间的有效采样的三维扩散磁共振技术，(2)获取 结合11.7T全脑标本的高空间-角分辨率3D dMRI数据，(3)到 用一种微分同胚非线性方法构建全海马区的计算平均Hardi图谱 基于旋转不变Hardi特征和黎曼平均的配准框架 研究正常衰老和AD患者海马区结构的区域性变化。这个项目带来了 结合创新的成像技术和计算分析方法，将使 具有显微结构细节的高场dmri图谱。建成后，这些目标将建立第一个哈迪 人类海马体图谱，这将是神经成像研究的宝贵资源，并提供 对促进我们对AD和衰老中的海马区病理的理解有重要的见解。

Summary The hippocampal formation is a structure of acute interest in neuroimaging, due to its crucial role in a wide spectrum of neurological disorders, including epilepsy, schizophrenia, and Alzheimer’s disease (AD). Three- dimensional mapping of the complex intra-hippocampal organization and subfield connectivity is of paramount importance, since different subregions of the hippocampus are known to exhibit selective and differential vulnerability to different neurological diseases, such as epilepsy and AD. Despite its central role in major neurological disorders, the internal laminar microstructure and subfield connectivity of the human hippocampus remain difficult to delineate with three-dimensional imaging techniques. The goal of this R01 proposal is to develop the first three-dimensional diffusion MRI computational atlas of the intact whole human hippocampus. The proposed atlas will build on key technical developments enabling combined high spatio-angular resolution diffusion MRI (dMRI) of the intact fixed hippocampus at 11.7T. The resulting atlas based on high angular resolution diffusion imaging (HARDI) data from multiple hippocampus specimens will provide delineation of human intra-hippocampal organization and circuitry in unprecedented microstructural detail. Further, current understanding of hippocampal circuitry and its disruption in disease is largely based on two-dimensional histological sections. The imaging platform and dMRI atlas established here will allow for the first time, investigation and three-dimensional mapping of region-specific changes in intra- hippocampal structure in normal aging and AD brains, providing details of hippocampal pathology in greater detail than previously possible. We will achieve these goals in four specific aims: (1) to develop accelerated three-dimensional diffusion MRI techniques for efficient sampling of the combined (k,q)-space, (2) to acquire combined high spatio-angular resolution 3D dMRI data of whole hippocampal specimens at 11.7T, (3) to construct a computational averaged HARDI atlas of the whole hippocampus using a diffeomorphic non-linear registration framework based on rotation-invariant HARDI features and Riemannian averaging, and (4) to investigate regional changes in intra-hippocampal structure in normal aging and AD. This project brings together innovative imaging techniques and computational analysis methods, that will allow development of the high-field dMRI atlas with microstructural detail. Upon completion, these goals will establish the first HARDI atlas of the human hippocampus, which will be an invaluable resource for neuroimaging studies, and provide significant insights to advance our understanding of hippocampal pathology in AD and aging.