Establishment of the Diagnostic Model in Solitary Pulmonary Nodule Appearing as Ground-glass Nodule / 中国医学影像学杂志

ABSTRACT

Purpose To explore the independent predictors of malignant solitary pulmonary nodule (SPN) manifesting as ground-glass nodule (GGN),and to establish a prediction model.Materials and Methods The clinical data and CT images of 362 patients (group A) with pathological-confirmed SPN appearing as GGN in Shanghai Chest Hospital Shanghai Jiaotong University from January 2014 to December 2015 were retrospectively analyzed.The independent predictors of malignant SPN were identified,and the clinical prediction model was established.Another 119 SPN patients in Affiliated Zhoushan Hospital of Wenzhou Medical University were selected as group B to verify the diagnostic efficiency of the prediction model.Results Using multivariate Logistic regression analysis,clear border (OR=6.274,P＜0.01),smooth edge (OR=0.391,P＜0.01),lobulation (OR=3.387,P＜0.01),pleural retraction sign (OR=2.430,P＜0.01),and vocule sign (OR=3.076,P＜0.01)were identified as independent predictors of malignant SPN.The area of the model under the ROC curve was 0.859 with 95％ CI (0.804-0.903).The diagnostic accuracy rate,sensitivity,specificity,positive predictive value and negative predictive value were 85.92％,91.03％,81.97％,92.03％ and 73.53％,respectively.Conclusion In this study,the independent predictors of malignant SPN appearing as GGN were identified,and the prediction model was established.The model can accurately identify SPN and provide effective help for early diagnosis of SPN.