| Objectives:To develop two combined clinical-radiomics models of pericoronary adipose tissue（PCAT）for the presence and characterization of non-calcified plaques on plain CT scan.Methods:Altogether,431 patients undergoing Coronary Computed Tomography Angiography（CCTA）in the Department of Radiology of the First Hospital of Jilin University from March 2019 to June 2021 who had complete data were enrolled,including 173 patients with non-calcified plaques of the right coronary artery（RCA）and 258 with no abnormality.PCAT was segmented around the proximal RCA on CT plain scan（calcification score sequence）.Two best models were established by screening features and classifiers respectively using Fe Ature Explorer software.Model 1 distinguished normal coronary arteries from those with non-calcified plaques,and model 2 distinguished vulnerable plaques in non-calcified plaques.Performance was assessed by the area under the receiver operating characteristic curve（AUC）.Results: 4 and 9 features were selected for the establishment of the radiomics model respectively through Model 1 and 2.In the test group,the AUC values,accuracy,sensitivity and specificity were 0.833,78.3%,80.8%,76.6% and 0.7467,75.0%,77.8%,73.5%,respectively.Gender,age,hypertension,diabetes,smoking,clinical diagnosis and mean CT value of coronal fat were independent risk factors for predicting the presence of non-calcified plaques,while hypertension was independent risk factors for predicting the presence of vulnerable plaques.The combined model including radiomics features and independent clinical factors yielded an AUC,accuracy,sensitivity and specificity of 0.896,81.4%,86.5%,77.9% for model 1 and0.752,75.0%,77.8%,73.5% for model 2,which were also higher than the fat attenuation index（FAI）model（AUC : 0.716,0.616）.Conclusion:The combined clinical-radiomics models based on plain CT images of PCAT had good diagnostic efficacy for non?calcified and vulnerable plaques. |
PDF Full Text Request