〔Abstract〕 Objective To explore the application value of CT threshold segmentation in the pathological classifi cation and infi ltration grade determination of ground glass nodules and some solid nodules in the lung. Methods Select 90 patients with subsolid nodules in the lung detected by the CT imaging department of Zengcheng branch of Nanfang hospital from January 2019 to November 2020, and divide them into low-invasive group with 50 cases and high-invasive group with 40 cases according to the pathological examination results, and compare the two groups The quantitative parameters of CT scans and their independent predictors in the level of invasion were analyzed, and the clinical value of CT threshold segmentation in judging the level of pathological invasion in patients with subsolid nodules in the lung was analyzed. Results Under different CT thresholds, when the threshold is -250 HU, the diagnostic effi ciency of subsolid nodules is the highest. The quantitative parameters of CT scan such as 1D-SCLW, 2D-SCLW, 1D-SCMW, 2D-SCMW, SCT in the high-invasive group were signifi cantly higher than those in the lowinvasive group (P < 0.001). Multivariate regression analysis of CT quantitative parameters found that SCT was an independent predictor of pathological infiltration of subsolid pulmonary nodules [OR = 1.164, 95% CI (1.137, 1.192)]. Conclusion The CT threshold segmentation method can accurately determine the type of subsolid pulmonary nodules, and perform quantitative volume analysis to determine that subsolid volume is an independent predictor of pathological infi ltration grade, and the larger the subsolid volume, the higher the degree of lung nodule infi ltration It has high clinical application value for pathological judgment and prognostic evaluation of subsolid nodules.