A Prediction Model For Benign And Malignant Solitary Pulmonary Nodules Based On The Maximum Diameter Of Nodule

Background and Objective: In this study,mathematical prediction equations were constructed with the maximum SPN diameter of 8 mm as the boundary,and a new SPN benign and malignant prediction model suitable for Chinese population was established..Methods: The clinical data and CT imaging data of 601 cases of SPN hospitalized in Qingdao Municipal Hospital Group from January 2017 to October 2020 were retrospectively collected and randomly divided into group A(501 cases)and group B(100cases)by Random number grouping method.Group A cases were divided into two parts(≤8mm and >8mm)by the maximum diameter.We used Single factor analysis and Logistic regression analysis to screen out independent influencing factors in order to establish a prediction model.Group B was selected for verification,and the model’s performance was compared with Mayo model and Peking University model with higher recognition degree.Results: According to the regression analysis,for the cases with maximum diameter≤8mm,smoking history,smooth edge,burr,calcification and vascular cluster sign were the independent influencing factors of benign and malignant of SPN.For cases with maximum diameter > 8mm,smoking history,tumor history and density,smooth edges,lobulation,calcification,pleural depression,vascular cluster sign and bronchial air sign were independent influencing factors of benign and malignant of SPN,and a prediction model was established.The data of group B were substituted into the model and the ROC curve was drawn.The AUC of this model was calculated to be 0.824 ± 0.044,which was better than Mayo model,and the difference was statistically significant,and which was better than Peking University model without the statistically significant difference.The sensitivity,specificity,positive predictive value and negative predictive value of this model for detecting benign and malignant of SPN were 82.2%(51/62),71.1%(27/38),80.9%(51/63)and 73.0%(27/37)when we used 0.683 as the cut-off point T.Conclusion: In this study,the mathematical prediction equations were constructed with the maximum diameter of SPN 8mm as the boundary layer.The established model has good performance in predicting the benign and malignant SPN,which is better than Mayo model.The difference is statistically significant,which is better than Peking University model,but the difference is not statistically significant.It can assist in the clinical diagnosis and management of solitary pulmonary nodules.The study found that for SPN with different maximum diameter,the independent predictors affecting its benign and malignant are completely different.Therefore,in clinical diagnosis and management of SPN,it is necessary to specify individualized diagnosis and treatment strategies for different patients.