Objective To investigate the predictive value of high-resolution CT quantitative parameters for the invasion of pure ground-glass nodules (pGGNs). Methods Preoperative CT quantitative parameters of 163 pGGNs lesions diagnosed by pathology in 148 patients were retrospectively analyzed, including maximum diameter, maximum vertical diameter, maximum cross-sectional area, volume, mass and mean CT value. According to the pathological classification, pGGNs were classified into two groups, the non-invasive group [including atypical adenomatous hyperplasia (AAH), adenocarcinoma in situ (AIS) and minimally invasive adenocarcinoma (MIA)], and the invasive group [only invasive adenocarcinoma (IAC)].The difference in CT quantitative parameters between the two groups of nodules was compared, the receiver operating characteristic (ROC) curve and logistic regression models were used to evaluate the predictive value of CT quantitative parameters on the degree of invasion of pGGNs. Results The maximum diameter, maximum vertical diameter, maximum cross-sectional area, volume, mass and average CT value of non-invasive group were significantly lower than those of invasive group (P<0.05). The predictive value of CT quantitative parameters from high to low were the maximum cross-sectional area (AUC=0.846), mass (AUC=0.834), volume (AUC=0.811), maximum diameter (AUC=0.803), maximum vertical diameter (AUC=0.799) and mean CT value (AUC=0.685), allwiththe statistical significance (P<0.05). The logistic regression analysis showed that the maximum cross-sectional area (OR=2.307, 95% CI: 1.689- 3.150, P<0.001) was the only independent predictor of the invasiveness of pGGNs with a predictive threshold of 2.224 cm2. Conclusion Preoperative CT quantitative parameters can effectively predict the invasion of lung pGGNs, and the maximum cross-sectional area of pGGNs has the highest predictive value.