Abstract: Objective: To explore the clinical value of a CT radiomics-based nomogram model in predicting the growth trend of pulmonary ground-glass nodules (GGNs). Methods: A retrospective study was conducted involving 340 patients with pulmonary GGNs identified on CT images from January 2014 to December 2024. Based on growth criteria, patients were divided into a stable group (224 cases) and a growth group (116 cases). Clinical data and imaging features were recorded. The dataset was randomly split into training and testing sets in a 7︰3 ratio. A logistic regression model combined 11 radiomic features selected using Lasso and clinical-imaging features filtered through univariate and multivariate analyses. Model performance, calibration, and clinical usefulness were assessed using the area under the curve (AUC), calibration curve, and decision curve analysis (DCA), respectively. Results: The AUC values (95% CI) for the training and testing sets were 0.950 (0.922-0.978) and 0.964 (0.926-1), respectively. The sensitivity, specificity, accuracy, and F1 score of the model were 0.889, 0.898, 0.895, and 0.852, respectively, for the training set and 0.857, 0.925, 0.902, and 0.857, respectively, for the testing set. The calibration curve and DCA showed that the model exhibited good fit and clinical practicality. Conclusion: The CT radiomics-based nomogram model can effectively predict the growth of pulmonary GGNs with stable and reliable predictive performance. The model can provide a reference for individualized follow-up management of GGNs in clinical practice. Routine follow-up protocols can be adopted for stable GGNs, whereas enhanced dynamic monitoring and timely evaluation of intervention indications should be implemented for GGNs with high growth risk, thereby offering a reliable imaging basis for precise diagnosis, treatment, and standardized management of GGNs.