Simultaneous imaging of myocardial blood flow and glucose metabolism using dynamic 18F-FDG PET

 描述(由申请人提供)：在美国大约有300万人患有缺血性心肌病。这种形式的心力衰竭是心肌梗死或严重的冠心病的结果，这些疾病降低了心脏的生存能力和功能。当存活心肌的患者没有血运重建时，缺血性心肌病与较差的长期存活率有关。通过对心肌血流和葡萄糖代谢的成像以及寻找血流-代谢不匹配，正电子发射断层扫描(PET)方法已被确立为评估心肌存活的金标准，以选择最能从外科血管重建术中获益的患者。目前的PET方法使用两个单独的静态扫描和两个不同的放射性示踪剂来产生流动-代谢图像对。虽然使用最广泛的放射性示踪剂18F-氟代脱氧葡萄糖(FDG)获得葡萄糖代谢的图像，但使用放射性示踪剂13N-氨或Rb-82进行心肌血流成像的临床应用有限。此外，两个独立成像会话的成像协议既耗时又耗费资源。因此，尽管PET的准确性很高，而且在过去十年中PET/CT扫描仪的安装迅速增加，但通过PET检测心肌存活目前在临床上仍未得到充分利用。在这个项目中，我们建议开发一种新的PET方法来评估心肌存活，该方法只使用一次FDG注射，而不需要特定的血流示踪剂。我们假设心肌血流量可以从动态FDG PET的定量动力学参数中得出。我们将开发一种新的多变量预测模型，使用统计机器学习来预测动态FDG PET数据中的心肌血流量。我们还将开发一种简化的动态FDG PET方案，以提高实用性。这一创新将提供血流-代谢图像对，用于在临床上有利的时间、成本和较低的辐射剂量下评估心肌存活。这项研究的成功将使PET心肌存活评估更广泛地应用于临床，与传统的两次会议方案相比，更容易获得，更低的辐射剂量，更便宜的成像成本和更短的临床就诊时间，从而改善我们治疗缺血性心肌病的临床实践。

 DESCRIPTION (provided by applicant): Ischemic cardiomyopathy affects approximately 3 million people in the United States. This form of heart failure is the result of myocardial infarcton or severe coronary heart disease that reduces the viability and function of the heart. Ischemic cardiomyopathy is associated with poor long-term survival when patients with viable myocardium are not revascularized. By imaging myocardial blood flow and glucose metabolism and seeking flow-metabolism mismatches, positron emission tomography (PET) method has been established as the gold standard of assessing myocardial viability for selecting patients who can benefit most from surgical revascularization. Current PET method employs two separate static scans with two different radiotracers for generation of the flow-metabolism image pair. While the image of glucose metabolism is acquired using the most widely used radiotracer 18F- fluorodeoxyglucose (FDG), myocardial blood flow imaging with the radiotracer 13N-ammonia or rubidium-82 suffers from limited clinical availability. In addition, the imaging protoco of two separate imaging sessions is time consuming and resource intensive. As a result, myocardial viability via PET is currently under-utilized in clinic despite its high accuracy and th fast-growing installation of PET/CT scanners in the past decade. In this project, we propose to develop a novel PET method for myocardial viability assessment that only uses a single injection of FDG without the need of a flow- specific radiotracer. We hypothesize that myocardial blood flow can be derived from the quantitative kinetic parameters of dynamic FDG PET. We will develop a new multi-variable prediction model using statistical machine learning to predict myocardial blood flow from dynamic FDG PET data. We will also develop a shortened dynamic FDG PET protocol to improve practicality. This innovation will provide the flow-metabolism image pair for myocardial viability assessment in a clinically favorable time, cost and with reduced radiation dose. Success of this research will make PET assessment of myocardial viability more widely available in clinic with easier access, lower radiation dose, cheaper imaging cost and shorter clinical visit time as compared with conventional two-session protocols, thus improving our clinical practice of treating ischemic cardiomyopathy.