Unbiased longitudinal neuromorphometry for clinical decision support

项目摘要 正常的人类神经解剖学是变化无常的，而且随着年龄的增长而增加。这阻碍了 神经成像检测阿尔茨海默病等神经疾病影响的能力 (AD)、亨廷顿病(HD)、多发性硬化症(MS)和精神分裂症。最近的大部分时间 现有的最先进的定量成像工具仍然使用横断面方法来分析 反复扫描。这些工具缺乏监控细微渐进变化的敏感度，因为 这些方法不能解释正常神经解剖学的巨大内在变异性。这个 该项目的目标是将一种纵向的神经形态图像处理商业化 用于放射学、神经病学和相关临床领域的管道。圆满完成 该项目将产生临床上有用的神经形态纵向分析流 比目前商业上可用的统计能力更强。权力的增加将 直接转化为增强的检测和评估个人进展的能力 和小组级别。它还将缓解当前纵向神经放射学中的一个主要痛点 阅读工作流程，缩短放射学报告周转时间(RTAT)。

Project Summary Normal human neuroanatomy is incredibly variable, and increases with age. This impedes the ability of neuroimaging to detect effects in neurological conditions such as Alzheimer's disease (AD), Huntington's disease (HD), multiple sclerosis (MS) and schizophrenia. Most of the recently available state-of-the-art quantitative imaging tools still use cross-sectional methods to analyze repeated scans. These tools lack the sensitivity to monitor subtle progressive changes because such approaches do not account for the large intrinsic variability of normal neuroanatomy. The goal of this project is to commercialize a longitudinal, neuro-morphometric image processing pipeline for use in radiology, neurology and related clinical fields. The successful completion of this project will result in a clinically useful neuro-morphometric longitudinal analysis stream with more statistical power than is currently available commercially. This increase in power will directly translate into an enhanced ability to detect and assess progression at both the individual and group levels. It will also alleviate a major pain point in current longitudinal neuroradiology reading workflows, reducing radiology report turnaround times (RTAT).