Abstract: Objective To analyze the predictive value of a multimodal CT radiomics model in predicting unstable angina in patients with aortic plaques. Methods A total of 236 patients with aortic plaque who underwent multimodal CT examination in our hospital from January 2023 to December 2023 were selected for this study. Combined with 1-year follow-up, they were divided into non-occurrence group (200) and occurrence group (36), and the difference data between the two groups were replaced by Logistic regression equation. The influencing factors of unstable angina pectoris were determined, the predictive model was constructed, and the receiver operating characteristic curve (ROC) was drawn to evaluate the model efficacy. Results The location of aortic plaque, degree of aortic stenosis, CT value of aortic plaque, calcification score of aortic root, and volume of aortic calcification were all influential factors for the occurrence of unstable angina pectoris in patients with major arterial plaque, and the OR values were all greater than 1. By constructing a nomogram, ROC curve was drawn to verify the prediction efficiency of the model and the AUC was found to be 0.990 (95%CI: 0.970-1.000). Conclusion The multi-modality CT image omics prediction model has a high value in predicting unstable angina pectoris in patients with aortic plaque, and can provide evidence for clinical screening of high-risk groups and taking measures.

Key words: aortic plaque, unstable angina pectoris, CT angiography, multi-slice spiral CT, spectral CT