The AIPLAQUE Study: An Artificial Intelligence-based Prospective Study to Analyze PLAQUE Using CCTA

Acute coronary syndrome (ACS) is one of the leading causes of coronary artery disease (CAD) death worldwide. Vulnerable plaque rupture is a primary underlying cause of luminal thrombosis responsible for provoking ACS. Therefore, identifying high-risk plaques before ACS occurs has been a major research goal and requires further clinical perspectives. Coronary computed tomography angiography (CCTA) is a comprehensive, non-invasive and cost-effective imaging assessment approach, which can provide the ability to identify the characteristics and morphology of high-risk atherosclerotic plaques associated with ACS. Optical coherence tomography (OCT) is a new, light-based, intravascular imaging technique that provides high-resolution, cross-sectional images of coronary artery anatomy. Due to its superior resolution, OCT is more accurate in measuring the sites of plaque vulnerability, distinguishing the differences in its composition, informing about the anatomic severity of epicardial stenoses, and also provides input for computational models to assess functional severity. The objectives of the study are: (1) To construct an artificial intelligence model for identifying coronary plaque components on CTA images using OCT as the reference standard. (2) To conduct fluid mechanics simulation including blood vessel wall and plaque by using geometric and physiological models of blood vessels and plaques, and to provide more accurate functional parameters (CT-FFR). The enrollment criteria will be (1) Patients who presented with stable angina pectoris or acute coronary syndrome; (2) patients who meet the indications for coronary CT angiography, percutaneous coronary angiography and intravascular imaging; (3) Among those patients, patients who have at least one coronary artery stenosis of 30% - 90% in diameter ≥ 2mm confirmed by CCTA. Data collected will include CCTA, full angiographic, and OCT images. Combined with CTA/ICA/OCT images of multiple modalities, this study will develop a novel images analysis technology to automatically extract vascular lumen, plaque characterization, fluid-solid mechanical properties, and myocardial ischemia conditions using computational fluid dynamics (CFD) simulation and artificial intelligence deep learning.