Objective To analyze the value of CT imaging features in the diagnosis of pulmonary pure ground glass nodule (pGGN) invasive adenocarcinoma. Methods 40 cases of invasive lung adenocarcinoma patients confirmed by surgical pathology and whose CT showed pGGN from 2015 to 2017 in our hospital were retrospectively analyzed, among them, there were 24 cases of micro invasive adenocarcinoma (MIA) and 16 cases of invasive adenocarcinoma (IAC), the imaging features such as lesion size, density, edge and adjacent structures of the lesions were observed by CT and the average CT value were calculated, receiver operational characteristic (ROC) curve was used to assess the values of lesions size and average CT value in the diagnosis of pGGN invasive lung adenocarcinoma. Results The presences of irregular shape, burr sign, lobulation sign, vascular cluster sign and air bronchial sign in IAC group were significantly higher than those in MIA group (P<0.05), the difference of pGGN lesion size and average CT value between the two groups was statistically significant (P<0.05). ROC curve analysis showed that the best limit value of lesion size for predicting IAC was 18.45mm, AUC was 0.808, sensitivity, specificity and Jordan index was 66.70%, 88.00% and 0.547, respectively; the best limit value of average CT value for predicting IAC was -534.05Hu, AUC was 0.856, sensitivity, specificity and Jordan index was 80.00%, 84.00% and 0.640, respectively, the diagnostic value of average CT value is higher than that of the lesion size. Conclusion CT imaging features can be used to identify MIA and IAC in the diagnosis of pulmonary pGGN invasive adenocarcinoma, the average CT value has higher diagnostic value, IAC is more likely when pGGGN is irregular, accompanied by burr sign, lobulation sign, vascular cluster sign and air bronchial sign, it can provide reference value for clinical diagnosis and treatment.