Quantitative Coronary Hemodynamics Derived from CT Angiography

描述(由申请者提供)：这个有指导的职业发展奖将使一个多产、创新、成功的初级科学家的培训和职业过渡成为可能。一个关键因素是心血管诊断成像和医学领域的思想领袖的培训和指导。申请者米苏拉斯博士曾在布朗和麻省理工学院接受过计算机科学、物理和应用数学方面的培训。他是一位成功的物理学家，对该领域做出了重要贡献。他是哈佛大学放射学助理教授，以及由雷布基博士指导的BWH应用成像科学实验室(AISL)的MR物理和工程学主任。作为一名成功的私人投资和熟练的导师，雷比基博士将热情地支持和指导米苏拉斯博士实现职业独立。冠状动脉疾病(CAD)患者的非侵入性成像正被CT所吸引；这项拟议的研究解决了它目前在临床护理中的一些局限性。AISL评估新的CT技术，并研究和发表了冠状动脉增强的潜力，以扩展临床信息。人们认识到，临床心脏CT的下一次飞跃是血流定量；米苏拉斯博士建议 对这一系列工作采取一种变革性的新方法。对于应聘者来说，从MR基础物理学到临床CT应用的过渡创造了大量未得到满足的培训需求。基于课堂的培训(哈佛/麻省理工学院健康科学与技术部)和基于实用成像的培训(BWH非侵入性心血管成像计划)将理想地弥合这一差距。这份K01提案还纳入了知名专家的指导意见，这些专家了解米苏拉斯博士并渴望与他们合作。该团队包括杰出的临床医生和导师梅尔·克劳斯博士、CT物理学专家潘小川博士、造影剂交付专家Tae Bae博士、心肌灌注建模专家Al Lardo博士和高级临床应用专家Joao Lima博士。这个杰出的团队将指导米苏拉斯博士进行一个结构化的研究项目，利用他们的集体专业知识和米苏拉斯博士的现有技能。总而言之，Tey将通过应用创新方法从临床CT血管成像中提取血流，从而扩大非侵入性CAD成像的作用。目前，对血流的非侵入性访问仅限于使用从临床图像分割的动脉腔来模拟流体动力学。我们用一种从基于CT的血流到CT衍生的血流的新的过渡来推进这一领域：定量血流直接或间接来自体内观察到的动脉对比增强的时空模式。越来越多的文献支持定性血流信息嵌入动脉的CTA增强模式。我们表明，这些模式实际上具有平滑的空间和时间变化，这些变化与流动有着内在的联系。通过将信号处理和模式识别与高性能流体动力学建模相结合，这些信息将首次被访问，以量化真实的特定于患者的血流，并产生特定于病变的血流动力学指标，如血流储备分数。

DESCRIPTION (provided by applicant): This mentored career development award will enable the training and career transition of a productive, innovative, successful junior scientist. A key element is training and mentorship by thought-leaders in the field of cardiovascular diagnostic imaging and medicine. The applicant, Dr. Mitsouras, has been trained in computer science, physics, and applied mathematics at Brown and MIT. He is a successful MR physicist with important contributions to the field. He is Assistant Professor of Radiology at Harvard, and the Director of MR Physics and Engineering of the BWH Applied Imaging Science Lab (AISL), directed by Dr. Rybicki. A successful PI and skilled mentor, Dr. Rybicki will enthusiastically support and mentor Dr. Mitsouras to career independence. Noninvasive imaging for patients with coronary artery disease (CAD) is gravitating to CT; the proposed research addresses some of its current limitations in clinical care. The AISL evaluates new CT technologies, and has studied and published the potential of coronary enhancement to extend clinical information. It is recognized that the next leap in clinical cardiac CT is flow quantification; Dr. Mitsouras proposes a transformative new approach to this body of work. For the candidate, the transition from MR basic physics to clinical CT applications creates a large, unmet training need. Classroom based (Harvard/MIT Division of Health Sciences and Technology) and practical imaging based (BWH noninvasive cardiovascular imaging program) training will ideally bridge this gap. This K01 proposal also integrates mentorship from renowned experts who know and are eager to work with Dr. Mitsouras. The team includes Dr. Mel Clouse, an exceptional clinician and mentor, Dr. Xiaochuan Pan, an expert in CT physics, Dr. Tae Bae, an expert in contrast delivery, Dr. Al Lardo, an expert in modeling myocardial perfusion, and Dr. Joao Lima, an expert in advanced clinical applications. This exceptional team will navigate Dr. Mitsouras in a structured research project, capitalizing on their collective expertise and Dr. Mitsouras' existing skills. Together, tey will extend the role of noninvasive CAD imaging by applying innovative methods to extract blood flow from clinical CT angiography. At present, noninvasive access to flow is limited to simulated fluid dynamics using an arterial lumen segmented from clinical images. We advance this field with a novel transition from such CT-based flow, to CT-derived flow: quantitative flow that is derived either directly or indirectly from the spatio-temporal patterns of arterial contrast enhancement observed in vivo. A growing body of literature supports qualitative flow information is embedded in CTA enhancement patterns across an artery. We show these patterns in fact possess smooth space and time variations that are intrinsically linked to flow. By unifying signal processing and pattern recognition with high-performance fluid dynamics modeling, this information will be accessed for the first time to quantify true patient-specific flow and yield lesion-specific hemodynamic metrics such as fractional flow reserve.