This study aimed to develop and validate a CT-based radiomic model for differentiating prognostic subtypes of lung adenocarcinoma. From 993 patients with invasive lung adenocarcinoma, predominant histologic subtypes were categorized into three prognostic groups: lepidic (group 0), acinar/papillary (group 1), and solid/micropapillary (group 2). Using contrast-enhanced CT scans, 718 radiomic features were extracted from segmented tumors, with segmentation performed using Aview software (Coreline Soft). A model-development set of 893 patients was created, reserving 100 image sets for testing. Feature selection was conducted using the least absolute shrinkage and selection operator method. The radiomic model achieved areas under the curve (AUCs) of 0.892 and 0.895 on the development and test sets, respectively. The model's test set accuracy was higher than the average accuracy of three radiologists (73.0% vs. 61.7%), with a notably higher positive predictive value for group 2. The study concludes that the CT-based radiomic model performs comparably to radiologists in classifying lung adenocarcinoma subtypes.