Precision Cardiac CT: Development of a Computational Platform for Optimizing Imaging

在美国，冠状动脉疾病(CAD)是主要的死亡原因。医学成像是 CAD的诊断和管理推动了新技术和新应用的发展。然而， 心脏成像仍然是一项具有挑战性的任务。总是有一定程度的时间模糊或运动 人工制品，其影响和记录的限制仍然不确定。此外，患者的身体习惯 而技术限制可能会导致高噪声和降低空间分辨率。用于成像设备 涉及像CT这样的电离辐射，辐射剂量也是一个需要最小化的无处不在的现实 而不会影响图像质量。临床试验是评价成像的最佳途径 技术，但不断扩大的技术和参数的数量使每一个应用程序都要进行试验 或者协议在实际和财务上都是不可行的。因此，医学成像研究人员、行业和 FDA正越来越多地转向计算机化的模拟或“虚拟试验”。 虚拟试验涉及使用计算工具完全在计算机上进行实验。现实 患者模型或幻影与经过验证的成像模拟相结合，以模拟成像检查 和病人的情况。这些随后可以用来确定不同的患者属性和 成像条件影响剂量、图像质量和预定义已知条件的描述。调查结果 可用于规定特定的成像协议和最佳扫描参数，这些扫描参数可定制为 个体患者解剖为有效的临床决策提供了足够程度的确定性。 该项目的目标是开发、验证并向研究社区分发计算的 平台(包括一系列具有真实有限元心脏模型的解剖可变模型， 现代成像设备的精确模型和一套图像质量指标)，以在 心脏动态成像。该虚拟框架可以扩展到任何数量的心脏疾病、成像 模式和技术。作为我们长期战略中的第一个案例研究，本项目的重点是CT 由于它既有很大的需求，也有很大的潜力为 计算机辅助设计的优化评价。如果CT图像质量不是最优的，评价CAD，尤其是程度 狭窄和高危斑块特征的特征，可能会受到影响。我们开发的工具将 提供第一个实际平台，以表征CT技术方面对图像的精确影响 在广泛的患者解剖结构上的质量，以实现心脏的最佳可视化 对于给定的患者，在可能的最低辐射剂量的条件下。这种方法有很大的潜力来 显著改善心脏病的临床研究，超越CT成像和CAD，铺平道路 更快地将新的心脏成像技术转化为临床并更精确和 个性化的患者管理。

Coronary artery disease (CAD) is the leading cause of death in the US. Medical imaging is integral to the diagnosis and management of CAD fueling the development of new technologies and applications. However, imaging of the heart continues to be a challenging task. There is always a degree of temporal blur or motion artifact, the impact and documented limitations of which remain uncertain. Additionally, patient body habitus and technical limitations may contribute to high noise and degrade spatial resolution. For imaging modalities involving ionizing radiation like CT, radiation dose is also an ever-present reality that needs to be minimized without compromising image quality. Clinical trials are the best avenue for the evaluation of imaging technologies, but the ever-expanding number of technologies and parameters make a trial for every application or protocol unfeasible, pragmatically and financially. As a result, medical imaging researchers, industry, and the FDA are increasingly moving toward computerized simulations or `virtual trials'. Virtual trials involve the use of computational tools to perform experiments entirely on the computer. Realistic patient models or phantoms are combined with validated imaging simulations to emulate imaging examinations and patient conditions. These can subsequently be used to ascertain how differing patient attributes and imaging conditions impact dose, image quality, and depiction of pre-defined known conditions. The findings can be used to prescribe specific imaging protocols and optimal scan parameters that are customized to individual patient anatomy to provide a sufficient degree of certainty for effective clinical decision-making. The goal of this project is to develop, validate, and distribute to the research community a computational platform (including a series of anatomically variable phantoms with realistic finite-element cardiac models, accurate models for modern imaging devices, and a suite of image quality metrics) to perform virtual trials in dynamic cardiac imaging. The virtual framework can be extended to any number of cardiac conditions, imaging modalities, and technologies. As a first case study in our long-term strategy, the focus of this project is on CT as it has both a great need for and great potential to provide high spatial and temporal resolution for the optimized evaluation of CAD. If CT image quality is not optimal, the evaluation of CAD, particularly the degree of stenosis and characterization of high-risk plaque features, may be compromised. The tools we develop will provide the first practical platform to characterize the precise impact of the technical aspects of CT on image quality over a wide range of patient anatomies with the view to enable optimal visualization of cardiac conditions at the lowest possible radiation dose for a given patient. The approach has great potential to significantly improve clinical investigations of heart disease, extending beyond CT imaging and CAD, paving the way towards faster translation of new cardiac imaging technologies into the clinic and more precise and personalized patient management.