The combined model integrating clinical, CT-derived imaging, and PCAT radiomic features (Model 4) predicted major adverse cardiovascular events with AUCs up to 0.854 in diabetic and 0.803 in non-diabetic patients, outperforming models lacking full feature integration.
Observational (n=1,000)
No
Does a predictive model integrating clinical factors, CT-derived parameters, and PCAT radiomics improve the prediction of MACE in patients with and without diabetes compared to clinical or imaging factors alone?
Integrating PCAT radiomics with CT-derived parameters and clinical risk factors significantly improves the prediction of long-term major adverse cardiovascular events in both diabetic and non-diabetic patients with coronary artery disease.
Effect estimate: Model 4 AUC 0.854 (95% CI 0.779–0.929) in diabetic patients training set; 0.706 (95% CI 0.578–0.833) testing set; Model 4 AUC 0.803 (95% CI 0.756–0.850) in non-diabetic training set; 0.705 (95% CI 0.6377–0.774) testing set
p-value: p < 0.05 for superiority of Model 4 over other models
Objective To compare the application value differences of PCAT radiomic features, clinical risk features and computed tomography (CT)-derived parameters in predicting Major adverse cardiovascular events (MACE) in patients with/without diabetes. Methods Retrospective analysis included 1,000 coronary atherosclerosis patients undergoing Coronary CT angiography (CCTA) (with/without diabetes: 274/726) from the Eighth Affiliated Hospital of Southern Medical University. Clinical/CT data were collected, extracting 285 PCAT radiomic features from three major coronaries. Least absolute shrinkage and selection operator regression identified MACE-associated radiomic features. Patients underwent random 6:4 training/testing cohort split. Four predictive models were constructed: Model 1 (clinical factors), Model 2 (imaging factors), Model 3 (imaging-radiomic features), Model 4 (all factors). Results In the training set, Model 4 showed the best performance: The area under the curves (AUC) of 0.803 95% confidence interval (CI): 0.756–0.850 and 0.854 (95% CI: 0.779–0.929) for groups with/without diabetes, respectively. Model 3 outperformed Model 2 in patients without diabetes ( p 0.05), but not significantly in diabetic patients ( p 0.05). Conclusion PCAT radiomics, CT-derived parameters, and plaque features demonstrate differential predictive value for MACE in patients with/without diabetes. Combining these with clinical risk factors provides most effective model for both.
Chen et al. (Fri,) conducted a observational in Adults (≥18 years) with coronary artery disease and calcification undergoing coronary computed tomography angiography, stratified by diabetes status (274 with diabetes, 726 without diabetes) (n=1,000). Predictive risk models integrating clinical factors, CT-derived imaging parameters, and pericoronary adipose tissue radiomic features vs. Risk models using clinical factors alone, or clinical plus imaging factors excluding radiomic features, or imaging plus radiomic features excluding clinical factors was evaluated on Major adverse cardiovascular events (MACE) defined as composite of cardiac death, congestive heart failure, rehospitalization for unstable angina, non-fatal myocardial infarction, or coronary revascularization (Model 4 AUC 0.854 (95% CI 0.779–0.929) in diabetic patients training set; 0.706 (95% CI 0.578–0.833) testing set; Model 4 AUC 0.803 (95% CI 0.756–0.850) in non-diabetic training set; 0.705 (95% CI 0.6377–0.774) testing set, p=p < 0.05 for superiority of Model 4 over other models). The combined model integrating clinical, CT-derived imaging, and PCAT radiomic features (Model 4) predicted major adverse cardiovascular events with AUCs up to 0.854 in diabetic and 0.803 in non-diabetic patients, outperforming models lacking full feature integration.