CT Quantitative Parameter Study Of Invasive Nature Of Pulmonary Ground Glass Nodules

Objective:To evaluate the predictive value of CT quantitative parameters for the invasion of GGNs.Methods:Our study included imaging date of 43 patients with ground-glass nodules detected on CT scans(50 nodules in total)between January 2018 and January 2020,all the patients have pathologically confirmed to be lung adenocarcinoma by surgical pathology.According to pathological results,the case were divided into group A and group B for comparative analysis.Group A was an non-invasive adenocarcinoma group,and group B was an invasive adenocarcinoma group.Group A includes AAH and AIS.Group B includes MIA and IAC.The CT imaging of selected cases were confirmed and then putted all images(original images and MPR images)of the 50 cases into the picture archiving and communication systems.To obtain the length of lesion,the width of the lesion,the height of the lesion,the mean CT attenuation.Then volume,mass,the two-dimensional mean size and the three-dimensional mean size were calculated for each lesion.All the date were analyzed by SPSS statistical software.The parameter information of lesions was analyzed by single factor analysis between the two groups.The difference between two groups was tested by rank sum test.Characteristic with a P value of less than 0.05 at univariate analysis were incorporated into binary logistic regression analysis,to identify the risk factors for distinguishing the invasiveness of the lesions.The receiver operating characteristic curve analysis revealed that the optimal cut-off of invasive adenocarcinoma.Result:50 cases of ground-glass nodules included 3 cases of typical adeno matous hyperplasia(6%),21 cases of adenocarcinoma in suit(42%),10 cases of m inimally invasive adenocarcinoma(20%),16 cases of invasive adenocarcinoma(32%).Between A and B group were analyzed quantitatively,the results showed t hat the length(P=0.016),the width(P=0.002),the height(P=0.029),the mea n CT attenuation(P=0.034),the two-dimensional mean size(P=0.005),the three-dimensional mean size(P=0.008),volume(P=0.009),mass(P=0.007).The difference was statistically significant.Then,significant quantitative parameters were inclu ded in multivariate regression analysis.Binary logistic regression analysis show ed that mass was independent risk factors for invasiveness of pulmonary groun d-glass nodules(OR=4.611,95%IC=(1.198～17.746,P=0.026).Using ROC cur ve and mass as predictive index,invasiveness of pulmonary ground-glass nodul es were predicted.The results showed that the optimal cut-off of mass was 0.561g,the area under the curve was 0.723,the sensitivity was 53.80%and the spe cificity was 91.70%.Conclusion:1.CT quantitative parameters can provide some reference for the determination of the invasive nature of ground-glass nodules.2.Quality parameters of the most predictive value.