QUALITATIVE HIGH FIELD STRUCTURAL IMAGE ANALYSIS

这个子项目是许多研究子项目中的一个 由NIH/NCRR资助的中心赠款提供的资源。子项目和 研究者（PI）可能从另一个NIH来源获得了主要资金， 因此可以在其他CRISP条目中表示。所列机构为 研究中心，而研究中心不一定是研究者所在的机构。 该项目的总体目标是开发和建立在我们早期资助的1.5T区域形态学分割和空间归一化工作的基础上，利用更高场成像中可用的新型成像对比度。 目标1：多序列融合和高分辨率组织分割：开发自动组织分割方法，该方法通过多采集变量T1加权成像数据，逐个体素估计占用率和组织类型概率的体素图。这将包括帧间协同对准、运动帧抑制、图像的强度不均匀性校正。 帧，以及该数据与典型研究中采集的其他MRI序列（如DTI和FLAIR）的共对齐和融合。 目标二：改进的全脑空间归一化：开发和完善用于新型图像对比度的多序列精细尺度空间归一化方法，该方法利用高各向同性分辨率的多重对比度。这将采用由基于鲁棒区域化熵的配准准则驱动的大规模同构和对称图像配准。我们还将构建和使用高分辨率 用于空间标准化靶模板。 目标3：老年成人脑中T1值的高分辨率图：使用目标1和目标2中开发的组织分割和空间归一化方法，构建基于人群的统计图，捕获一组正常对照受试者的预期T1值及其位置方差。 2，连同本资源建议书中制定的采集和重建方法。该统计图将在网上公布，供其他研究人员使用。

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. The overall aim of this project is to develop and build on our earlier funded work on the segmentation and spatial normalisation for regional morphometry at 1.5T, to make use of novel new imaging contrasts becoming available in higher field imaging. Aim 1: Multi-Sequence Fusion and High Resolution Tissue Segmentation: Development of automated tissue segmentation methods which estimate a voxel by voxel map of both the fractional occupancy and the probability of tissue types from multi-acquisition variable T1 weighted imaging data. This willincorporate inter-frame co-alignment, motion frame rejection, intensity inhomogeneity correction of the frames, and co-alignment and fusion of this data with other MRI sequences (such as DTI and FLAIR) acquired in a typical study. Aim 2: Improved Whole Brain Spatial Normalisation: Develop and refine multi-sequence fine-scalespatial normalisation methods for novel image contrasts, which make use of multiple-contrasts at high isotropic resolution. This will employ large scale diffeomorphic and symmetric image registration driven by robust regionalized entropy based registration criteria. We will also construct and use a high-resolution target template for spatial normalisation. Aim 3: High Resolution Map of T1 values in the Aging Adult Brain: Construction of a population based statistical map capturing the expected T1 values and their variance with location over a set of normal control subjects using the tissue segmentation and spatial normalisation methods developed in aims 1 and 2, together with the acquisition and reconstruction methods developed in this resource proposal. This statistical map will be published on the web as a resource for other researchers to make use of.