ObjectiveTo assess the accuracy of CT features of lung nodules (≤3 cm) in predicting the accuracy of the pathological subtype and degree of infiltration of adenocarcinoma. Methods We retrospectively analyzed the clinical data of 333 patients with non-cavitary pulmonary nodules diagnosed as adenocarcinoma by surgery and pathology in the China-Japan Friendship Hospital from 2011 to 2018, including 108 males and 225 females, aged 16-82 (59.57±10.16) years. The basic clinical data and CT characteristics of the patients were recorded. ResultsWhen the average CT value was ≥−507 Hu, the maximum diameter of the lung window was ≥14.5 mm, and the solid component ratio was ≥5.0%, it indicated more likely the invasive adenocarcinoma (IAC). The higher the average CT value of the nodule, the larger the maximum diameter of the lung window, and the more solid components, the higher the degree of infiltration. CT morphological features (including burrs, lobes, vascular signs, bronchial signs, pleural stretch or depression signs) were more common in IAC. Among them, burrs were more common in acinar adenocarcinoma and papillary adenocarcinoma. In invasive adenocarcinoma, the higher the risk of recurrence of the pathological subtype, the greater the average CT value. When the average CT value of IAC was >−106 Hu, and the proportion of solid components was ≥70.5%, the histological subtypes were more inclined to micropapillary/solid predominant adenocarcinoma. Conclusion The evaluation of CT features of lung nodules can improve the predictive value of histopathological types of lung adeno- carcinoma, thereby optimizing clinical treatment decisions and obtaining more ideal therapeutic effects.