Abstract: Objective: We aimed to develop a predictive model for lung adenocarcinoma with ground glass nodules (GGNs) based on artificial intelligence (AI) and computed tomography (CT) features, and to evaluate the model’s predictive value. Methods: A total of 261 GGNs from 225 patients diagnosed after surgery at our hospital were retrospectively collected and randomly divided into a training set and a validation set in an 8:2 ratio. The GGNs were classified into preneoplastic lesions and adenocarcinoma groups based on pathological results. AI-derived quantitative parameters and CT signs from the training set were compared between the two groups, and independent risk factors were identified using multivariate logistic regression. A predictive model and nomogram were developed, and model performance was assessed through the area under the ROC curve (AUC), calibration curve, and clinical decision curve analysis (DCA). The model was subsequently validated using the validation set. Results: Kappa test indicated good agreement between the two attending physicians in their assessment of CT signs. Baseline analysis revealed no statistical differences between variables in both training and validation sets. In the training set, lobulation sign (OR=3.147, 95% CI: 1.303-7.601), vacuole sign (OR=2.563, 95% CI: 1.109-5.922), vascular abnormalities (OR=3.551, 95% CI: 1.545-8.164), long diameter (OR=1.154, 95% CI: 1.014-1.312), and mean CT value (OR=1.006, 95% CI: 1.003-1.009) were identified as independent risk factors for adenocarcinoma in GGN after univariate and multivariate analysis. The predictive model constructed based on this information showed good discrimination ability, with an AUC of 0.901 (95% CI: 0.859-0.943) in the training set and an AUC of 0.896 (95% CI: 0.810-0.983) in the validation set, significantly outperforming individual risk factors. The Hosmer-lemeshow test demonstrated good model fit in both sets and DCA showed its strong clinical applicability. Conclusion: The model based on AI and CT signs demonstrated good predictive performance for GGN lung adenocarcinoma, providing valuable insights for clinical decision-making.