Development of a predictive model and nomogram using CTPA parameters for the identification of group 1 pulmonary arterial hypertension

This study aimed to identify computed tomography pulmonary angiography (CTPA)-derived predictors of pulmonary arterial hypertension (PAH) and to develop a prediction model and nomogram for non-invasive diagnosis. We retrospectively analyzed CTPA datasets from 203 patients in the training cohort and 98 patients in the validation cohort, stratified into PAH (confirmed by right heart catheterization) and non-PAH groups. Pulmonary vascular and cardiac structural metrics were measured. The least absolute shrinkage and selection operator (LASSO) and logistic regression identified interventricular septal angle (IVSA), interventricular septal thickness (IVST), and right ventricular outflow tract wall thickness (RVOTW) as independent predictors. The model achieved an area under the receiver operating characteristic curve (AUC) of 0.964 and 0.969 in the training and validation cohorts, respectively, with excellent calibration and clinical utility. The CTPA-based predictive model and nomogram may serve as a useful reference tool, potentially assisting in the identification of PAH and optimizing clinical workflows.