Risk Evaluation by COronary Imaging and Artificial intelliGence Based fuNctIonal analyZing tEchniques - III

This is a single-center, prospective cohort study. This study is designed to accurately analyze coronary artery plaque characteristics and local hemodynamic parameters in patients diagnosed with chronic coronary syndrome (CCS) or non-ST-segment elevation acute coronary syndrome (NSTE-ACS), with marginal lesions (diameter stenosis \[DS\] between 40%-69%) or obstructive lesions (DS ≥70% or CT-FFR \<0.8) in major coronary arteries by multimodality imaging including noninvasive coronary CT angiography (CCTA) and intracoronary imaging techniques, such as intravascular ultrasound (IVUS), optical coherence tomography (OCT), and near infrared spectroscopy (NIRS). The congruity and incongruity between different imaging modalities will be evaluated. The purpose of this study is to improve the accuracy and depth of plaque analysis by CCTA under the guidance of intracoronary imaging by co-registration, feature extraction, radiomic analysis and machine learning. Afterwards, a new CCTA-based high-risk plaque model will be constructed through the training process guided by intracoronary imaging and hemodynamic evaluation. The cohort will be followed-up every 3 months for 2 years. Cross-validation will be performed to compare the new model with traditional CTA models in detecting high-risk plaques and predicting the occurrence of major adverse cardiovascular events (MACEs). All the patients with CCS or NSTE-ACS, who undergo CCTA to confirm the presence of marginal or obstructive coronary lesions, and the subsequent invasive coronary angiography and intracoronary imaging examination within 1 month after CCTA will be enrolled.