Novel integration of coronary atherosclerosis imaging with physiologic features for optimized risk stratification of patients with coronary artery disease.

Background:

This project seeks to integrate coronary atherosclerosis with coronary pathophysiologic features to improve identification of stable individuals who will subsequently experience acute coronary syndromes (ACS). These determinations will take place in the largest consecutive cohorts of patients undergoing coronary computed tomography angiography (CCTA) to date, using comprehensive image, clinical and laboratory evaluation for coronary artery disease (CAD), rendering this proposal the most thorough effort to characterize CAD in stable individuals without known CAD who will experience future ACS.

Aims:

The project aims at integrating coronary atherosclerotic and physiologic features for prediction of ACS. These features will be studies by developing innovative technological approaches applying machine learning to CCTA that will provide unique insights into coronary atherosclerosis and pathophysiology, and integrates them into a clinical workflow after validation.

Workplan:

To achieve this level of innovation, we necessarily bring together an exceptional multidisciplinary group of investigators with unparalleled expertise whose unique perspectives will be synergistic for the successful completion of this project. Our team consists of expertise in CAD; CCTA; computational fluid dynamics (CFD); computer vision; machine learning; and clinical trials.

Our joint expertise and knowledge is fundamental for this project. The project spans over a period of three years. The first part aims at characterizing coronary pathophysiologic characteristics (CPCs) associated with future ACS.

The second part applies novel machine learning frameworks to integrate CPCs with atherosclerotic anatomic coronary plaque characteristics (APCs) for enhanced identification of stable individuals who will experience future ACS. The developed machine learning tool will be validated in a unique cohort of stable individuals with suspected CAD, allowing us to determine the primary role of CPCs to patient-centered outcomes.

Significance:

The project provides the rationale for a novel diagnostic and prognostic paradigm that can be readily applied in clinical care of patients with suspected CAD. Further, this work will offer unique insights into the pathophysiology of CAD.