Abstract Purpose To develop a quantitative image analysis for detecting coronary artery stenosis, calcification, and vulnerable plaque in reconstructed coronary computed tomography angiography. Materials and Methods 2,978 reconstructed coronary artery CT angiography (CCTA) of our hospital in South Korea from 2014 to 2022 were enrolled. For each CCTA scan, a total of 13 images were reconstructed for each coronary artery (CA) by CT Cardiac Package, TeraRecon Inc. A total of 769 CT images were independently selected as the test data set in the last period, while the remaining were utilized in a 4:1 ratio as train and validation datasets. The locations of stenosis and vulnerable plaques were strongly labeled by an expert radiologist with more than 10 years of experience to train a 2D segmentation model (SM). First of all, the correlation between a quantitative stenosis index (QSI) and conventionally measured quantitative angiographic analysis (QCA) was evaluated. We developed an SM for CA and stenosis using 2D nnU-Net. Using SM, the position and length of stenosis per image were evaluated, and a CA diameter profile was analyzed through Full Width at Half Maximum analysis. The QSI is acquired through the calculation of the CA diameter profile. Secondly, images with marked calcifications and calcification scores (CS) were evaluated by an expert cardiologist with more than 15 years of experience. The CS was calculated by measuring the number of voxels over 1100 Hounsfield units in the CA using the previously developed vascular SM. The result of CS was evaluated by visual scoring of the expert cardiologist with a 10-scale. Lastly, the presence or absence of vulnerable plaques per CA was evaluated. In addition, another SM for normal CA and vulnerable plaques were trained. For the evaluation method, sensitivity and specificity were calculated for each CA. Results The correlation between QSI and QCA was 0.715. Visual scoring on the result of CS was 7.150 ±2.354 to 10. The sensitivity and specificity of the vulnerable plaque for evaluating vascular unit classification were 0.929 and 0.910, respectively. Conclusions We developed an automated quantitative image analysis for evaluating QSI, detecting calcification, and vulnerable plaque in CCTA, which could be used in the actual screening setting.