Dynamic MRI for Myocardial Perfusion and Viability

描述（由申请人提供）：本提案旨在提高使用MRI表征缺血性心脏病的准确性和可重复性。前两个目标集中在动态MRI灌注：（1）开发径向k空间采集方法，提供快速测量钆浓度，以改善绝对心肌灌注的估计。(2)旨在改进和自动化动态数据的后处理分析，以获得稳健的定量灌注估计。将开发用于配准、分割和动力学建模的新方法，以自动生成绝对灌注图沿着置信图。(3)使用3 T MRI确定新灌注方法的准确性和可重复性。还将评估用于应变和瘢痕映射的新方法的可重复性。(4)将这些方法应用于接受骨髓干细胞治疗作为冠状动脉旁路移植术辅助治疗的患者的纵向研究。研究方法：将使用逼真的计算机模拟、体模和人体研究开发径向灌注序列，以提供高剂量动态MRI研究的准确钆浓度。将开发切片跟踪采集和组合配准/分割方法，以便对数据进行稳健的自动化处理。将开发不同的生理模型，并比较其提供绝对灌注值的能力。将开发和测试通过联合估计组织和动脉输入函数模型拟合来改善输入函数的数学方法。新的径向切片跟踪采集和处理方法将通过在相同患者中与动态PET定量灌注进行比较沿着与笛卡尔采集进行比较来验证。将在20名受试者中表征灌注方法以及应变和瘢痕标测技术的可重复性，每名受试者成像两次。然后，新方法将被应用于对干细胞治疗的心肌反应产生新的见解。该项目的成果将是用于心脏MRI研究的有效成像方案和软件，用于体内心肌灌注、应变和瘢痕纵向评估的改进方法，以及关于干细胞对心脏灌注、应变和瘢痕影响的新信息。这种方法将是非常宝贵的评价新的治疗和缺血性心脏病的检测和表征。

DESCRIPTION (provided by applicant): This proposal seeks to improve the accuracy and repeatability of the characterization of ischemic heart disease using MRI. The first two aims focus on dynamic MRI for perfusion: (1) To develop radial k-space acquisition methods to provide rapid measurements of gadolinium concentration to improve the estimation of absolute myocardial perfusion. (2) To improve and automate post-processing analysis of the dynamic data to obtain robust quantitative perfusion estimates. New methods for registration, segmentation, and kinetic modeling will be developed to automatically produce maps of absolute perfusion along with a confidence map. (3) To determine the accuracy and repeatability of the new perfusion methods using 3 Tesla MRI. The repeatability of new methods for strain and scar mapping will also be assessed. (4) To apply the methods to longitudinal studies of patients undergoing bone marrow stem cell therapy as an adjunct to coronary artery bypass grafting. Methods: Development of radial perfusion sequences to provide accurate gadolinium concentrations from high dose dynamic MRI studies will be pursued using realistic computer simulations, phantoms, and human studies. Slice tracking acquisitions and combined registration/segmentation methods will be developed for robust automated processing of the data. Different physiological models will be developed and compared in their ability to provide absolute perfusion values. Mathematical methods for improving input functions by jointly estimating tissue and arterial input function model fits will be developed and tested. The new radial slice-tracking acquisition and processing methods will be validated by comparison in the same patients to quantitative perfusion from dynamic PET, along with comparison to a Cartesian acquisition. The repeatability of the perfusion methods and the strain and scar mapping techniques will be characterized in 20 subjects, each imaged twice. The new methods will then be applied to generate new insights into the myocardial response to stem cell therapy. The outcome of this project will be validated imaging protocols and software for use with cardiac MRI studies, improved methods for longitudinal assessment of myocardial perfusion, strain, and scar in vivo, and new information regarding the effect of stem cells on cardiac perfusion, strain, and scar. Such methods will be invaluable for evaluating novel therapies and for the detection and characterization of ischemic heart disease.