Abstract: Objective: To establish a multi-parameter diagnostic model to determine lung adenocarcinoma with an invasive tendency, based on 3D CT quantitative parameters of subsolid nodules(SSNs), blood tumor markers and blood routine parameters. Materials and methods: One hundred and seven patients were retrospectively included, who had thin-slice CT scan, postsurgery histological examination, and blood tumor markers, and routine blood tests. The evaluated parameters included age and gender of patients, the maximum diameter, total volume, proportion of solid components, and average CT value of SSN in CT 3D-reconstruction, as well as blood laboratory indicators: CEA, CYFRA21-1, NSE, CA125, CA153, CA242, CA199, CA724, SCC, CRP, WBC, NEUT, and NEUT%. A multiple logistic regression model was established, and the area under the receiver operating characteristic curve(AUC) was used to evaluate the diagnostic capability of the model for SSN with an invasive tendency. Results: There were significant differences in age, maximum SSN diameter, total volume, the proportion of solid components, and mean CT values between the benign and preinvasive lesion groups(51 cases) and minimally invasive and invasive adenocarcinoma groups(56 cases)(P<0.05). In the multiple regression model established by the above parameters, the maximum diameter of SSN(P=0.007) and the proportion of solid components(P=0.004) were significant. The AUCs of maximum diameter, the proportion of solid components and the regression model to determine SSNs with an invasive tendency were 0.764, 0.749, and 0.801, respectively. Conclusion: In terms of CT quantification and some blood tumor markers and blood routine parameters that can be obtained in health check examination, the establishment of a comprehensive diagnostic model using SSN maximum diameter and proportion of solid components can effectively predict the invasiveness of SSN, which is helpful for the detection of patients requiring surgery in lung cancer screening.