Coronary artery disease (CAD) is among the leading cause of death and disability. Identification of patients at high risk of cardiovascular events is pivotal. However, current risk stratification based on imaging and known biomarkers is suboptimal. The objective of this proposal is to develop a multicriteria decision model for non-invasive assessment of vulnerable atherosclerotic patients and to evaluate its ability to predict the occurrence of an adverse event in intermediate-to-high risk patients with suspected or known CAD. The planned workflow includes a first step using a retrospective cohort of patients undergoing clinically indicated coronary angiography (CCTA) to develop an integrated application for automatic coronary artery segmentation, quantitative plaque analysis, biomechanics and fluid dynamics, based on machine learning, radiomics and computational analysis approaches and validated against the reference standard for each tool. The second step will apply this new methodology to a larger retrospective cohort of patients with the integration of genomic biomarker assessment to derive the most accurate risk stratification model to properly identify vulnerable patients and vulnerable plaques with respect to outcome. Finally, in the third step, the derived predictive model will be prospectively validated in an independent cohort of patients from an ongoing study (CTP-PRO study) to assess the robustness and accuracy of the proposed solution.