Mathematical/Computational Tools for Mapping Brain Data Bases

现代全脑冷冻切片和数字化技术与现代组织切片和染色方法相结合，使解剖学家能够提供关于个体大脑解剖结构和功能的精细详细的体内信息。难点在于两个方面：同一解剖结构内的图像和切片必须进行注册。但更根本的是，即使登记，不同解剖结构之间的正常变异也会使解剖结构间数据的汇集变得困难，如果不是不可能的话。在本提案中，我们致力于开发算法工具，用于通过猕猴大脑的全脑冷冻切片（0.05-0.1 mm分辨率）和染色组织切片测量的解剖和功能数据的融合。这一建议的主要焦点是_的发展，如在成年猕猴中表现出来的，允许将单个图谱映射到多个单独的目标。该图谱包含了在我们的工作中对研究者具有生物学意义的所有亚卷，包括脑沟和脑回、丘脑深核、脑室和枕叶的各种亚区，包括初级视觉皮层（V1区）和其他视觉区。我们的表征为从图谱（模板）及其子卷到目标解剖学家族的平滑地图（保留不同几何长度，面积，体积的地图）的算法生成提供了工具。这些地图允许从不同的实验准备中融合和汇集几何和生理信息到一个共同的坐标系统或数据库。作为正常猕猴解剖变异性表征的一部分，我们将从具有代表性的猕猴全脑解剖学家族中构建一个概率度量。概率测量了实际脑解剖中的正常解剖变异。这将为神经解剖学家提供一个直接的工具，将他们的样本放置在一个共同的坐标系中，具有精确的确定性或信心。这种解剖学变异的测量既反映了预定义的解剖学标记区域的潜在稳定性，即子区域和体积，如沟、eyri和不同的视觉区域，也反映了它们相对于单一生物学意义坐标系统的共变。

Modern whole brain cryosection and digitization techniques coupled to modern tissue section and staining methods allow anatomists to provide exquisitely detailed in vivo information regarding the anatomical structure and function of individual brains. The difficulty lies in two areas: images and sections within the same anatomy must be registered. But more fundamentally, even when registered, normal variation across disparate anatomes makes pooling of inter anatomical data difficult, if not impossible. In this proposal, we strive towards the development of algorithmic tools for the fusion of anatomical and functional data as measured via gross whole brain cryosections (0.05-0.1 mm resolution) and stained tissue sections in the macaque monkey brain. The major focus of this proposal is on the development of _ as manifest in the adult macaque, allowing for the mapping of a single atlas to multiple individual targets. The atlas contains all subvolumes of biological significance to the investigator in our work these would include the sulci and gyri, deep thalamic nuclei, ventricles and various sub areas in the occipital lobe, including primary visual cortex (area V1) and other visual areas. Our representations provide a tool for the algorithmic generation of smooth maps (maps preserving differential geometric lengths, areas, volumes) from an atlas (template) and its subvolumes to a family of target anatomies. These maps allow for the fusion and pooling of geometric and physiologic information from varied experimental preparations into one common coordinate system or data base. As part of such a representation of normal macaque anatomical variability, a probabilistic measure will be constructed from a family of representative macaque whole brain anatomies. The probability measures normal anatomical variation in act ual brain anatomy. This will provide neuroanatomists with a direct tool for placing their samples into a common coordinate system with a precise degree of certainty or confidence. Such a measure of anatomical variability reflects both the underlying stability of pre-defined anatomically labeled regions, i.e. subareas and volumes such as sulci, eyri, and different visual areas and their covariation relative to a single biologically meaningful coordinate system.