ML Model Sorts Multimodality Data Better, Predicts MACEs in Newly Diagnosed CAD

Background: Coronary artery disease (CAD) and cardiovascular disease are common causes of mortality worldwide, and risk stratification is important to enhance risk factor management and primary prevention. Coronary CT angiography (CCTA), noninvasive stress testing, stress cardiac MRI, and other modalities are often used to help assessment. Synthesizing data from multimodality imaging is challenging. Machine learning (ML) models, trained to discern patterns within extensive data sets with numerous variables, may assist in this data integration. Objective: To evaluate the performance of an ML model that uses both stress cardiac MRI and CCTA data to predict major adverse cardiovascular events (MACEs) in patients with newly diagnosed CAD. Design: Retrospective study. Methods: 2210 symptomatic patients without known CAD referred for CCTA between December 2008 and January 2020 were included. Patients with obstructive CAD at CCTA underwent stress cardiac MRI. Clinical, electrocardiogram, CC more...