Predictive Role Of Pathological Subtypes Of Ground-glass Nodular Lung Adenocarcinoma Based On Radiomics Features

Objective To verify the accuracy of the radiomics features model in identifying the pathological subtypes(infiltrating and non-infiltrating)of ground glass nodule(GGN)lung adenocarcinoma.Methods A retrospective collection of 180 surgical cases of General Thoracic Surgery,General Hospital of Ningxia Medical University,from January 2017 to November 2020,with GGN imaging and postoperative pathological confirmation of lung adenocarcinoma,a total of184 nodules,according to the ratio of 7:3,the nodules of the enrolled patients were randomly divided into training group and verification group,with 129 nodules in the training group and55 nodules in the verification group.The 1mm plain scan sequence in the CT image of ground glass nodules was selected to delineate the region of interest(region of interest,ROI),and 960 radiomics features were extracted from it,and 7 features were screened out by t-test and LASSO regression method.The support vector machine(SVM)method is used to construct the radiomics model in the training group and verify its accuracy in the verification group.The receiver operating characteristic(receiver operator characteristiccurve,ROC)curve was used to evaluate the predictive performance of the model on the pathological subtypes of GGN lung adenocarcinoma.Results The results of univariate analysis showed that there was no significant statistical difference between gender and lesion location and the pathological subtype of GGN(P>0.05),while the difference between age and the maximum diameter of nodule on CT and the pathological subtype of GGN was Statistically significant(all P<0.05).The AUC(95%CI)value of the radiomics model in the training group is 0.94(0.89～0.99),the accuracy is 91%,the sensitivity is 95%,and the specificity is 84%.In the validation group,the AUC(95%CI)value is 0.88(0.83～0.93),the accuracy is 83%,the sensitivity is 84%,and the specificity is 81%.Conclusion The model established based on the characteristics of radiomics has a predictive effect on the pathological subtypes of GGN lung adenocarcinoma,and can lay the foundation for the development of individualized diagnosis and treatment strategies for GGN lung adenocarcinoma.