Quantitative Magnetization Transfer Imaging for Early Detection of Alzheimer's Disease

项目摘要 阿尔茨海默病（AD）是痴呆症的最常见原因，并且目前是不可逆的神经退行性疾病， 生殖疾病在美国的疾病负担不断增加。临床管理和 AD的新疗法的开发将贝内于可重复的，稳健的和非侵入性的早期AD标记物。 体内AD病理学，但没有这样的工具是在常规临床实践中建立良好的。定量磁化 转移（qMT）是一种基于MRI的技术，用于检测微结构组织变化，例如脱髓鞘和 大分子的存在，包括淀粉样β蛋白。这种对艾米多方面的敏感性- 类蛋白/tau/神经变性框架提供了一种有前途的诊断替代淀粉样蛋白PET在一个单一的，快速 和高分辨率成像检查。我的初步发现表明，使用混合态自由理论， 在我的研究小组开发的旋进，qMT图像可以在12分钟内获得在体内在1 mm各向同性 分辨率高，SNR好。此外，通过与认知正常但淀粉样蛋白阳性的患者的PET直接比较， 受试者，qMT似乎对β淀粉样蛋白的早期积累敏感，并证实了先前的文献 AD中qMT汇率增加的报告。我的建议旨在解决剩下的技术挑战 围绕qMT，通过优化采集以提高对淀粉样蛋白β（Aim 1）的敏感性， 一个基于神经网络的重建管道，以大大减少后处理时间，并提高其 对磁场不均匀性的稳健性（目标2），这会在定量参数中产生偏差。这些 偏见是特别重要的削减在大脑深处，在那里的淀粉样蛋白β的积累据称 从公元开始。在目标3中，我将进行一项试点研究，以比较所提出的qMT方法的灵敏度， 将认知正常人群中的体内淀粉样蛋白β直接转化为淀粉样蛋白PET。总之，这将使qMT成为一个 体内淀粉样蛋白替代标记物并激发了对其在监测AD进展中的效用的进一步临床研究 和对治疗的反应

Project Summary Alzheimer's disease (AD) is the most common cause of dementia and, at present, an irreversible neurode- generative disease with an ever-increasing disease burden in the United States. Clinical management and de- velopment of novel therapeutics for AD will benefit from reproducible, robust and non-invasive markers of early in vivo AD pathology, but no such tools are well-established in routine clinical practice. Quantitative magnetization transfer (qMT) is an MRI based technique for detecting microstructural tissue changes such as demyelination and the presence of macromolecules, including amyloid beta protein. This sensitivity to multiple aspects of the amy- loid/tau/neurodegeneration framework offers a promising diagnostic alternative to amyloid PET in a single, rapid and high-resolution imaging exam. My preliminary findings suggest that, using the theory of hybrid state free precession developed in my research group, qMT images can be acquired in vivo in 12 minutes at 1mm isotropic resolution with good SNR. Additionally, by direct comparison to PET in a cognitively normal but amyloid positive subject, qMT appears to be sensitive to the early accumulation of amyloid beta and confirms previous literature reports of increased qMT exchange rates in AD. My proposal aims to solve the remaining technical challenges surrounding qMT by optimizing the acquisition for improved sensitivity to amyloid beta (Aim 1) and developing a neural network based reconstruction pipeline to substantially reduce the post-processing time and improve its robustness to magnetic field inhomogeneities (Aim 2), which create biases in the quantitative parameters. These biases are particularly important to curtail in the deep brain, where the accumulation of amyloid beta is purported to begin in AD. In Aim 3, I will perform a pilot study to compare the proposed qMT method's sensitivity for in vivo amyloid beta in a cognitively normal population directly to amyloid PET. Together, this will establish qMT as a surrogate marker for in vivo amyloid and motivate further clinical studies on its utility in monitoring AD progression and response to therapy.