Abstract: Objective To explore the impact of hidden lymph node metastasis of patients with thoracoscopy lung cancer,and build risk prediction model construction.Methods From April 2022 to August 2024,93 patients undergoing thoracoscopic radical resection of lung cancer were accepted.According to the case of hidden lymph node metastasis,the patient was divided into a group (n=28) and the uncomfortable group (n=65 ),Compare the clinical data of the two groups of patients,Logistic regression analysis affects the factors that occur in hidden lymph node metastasis,and use this to establish a risk predictive model.Internal verification,the consistency of the calibration diagram evaluation model is adopted.Results Multi-factor analysis shows that surrounding tumors,maximum diameter of tumor,≥2.5 cm,clinical staging in phase Ⅲ,tumor components are solid,burr signs,and vascular infiltration is hidden in patients with thoracoscopy lung cancer root treatment Risk factor for lymph node transfer (P<0.05).The ROC curve results showed that this model had a curve of hidden lymph node metastasis in patients with thoracoscopy lung cancer root treatment,with 0.906 and 95%CI:0.829~0.983,85.72%sensitivity,and 90.75% of the specific degree.The calibration diagram of the bootstrap method verification predictive model shows that the calibration curve is close to the ideal curve,the number of self -sampling=1 000,and the average absolute error=0.043.Conclusion The maximum diameter of the surrounding tumor,the maximum diameter of the tumor,the clinical stage of the clinical stage Ⅲ,the tumor component is solid,the burr signs,and the vascular infiltration is the risk factors of hidden lymph node metastasis in patients with thoracoscopy lung cancer.The risk prediction model established accordingly has high accuracy and reliability.

Key words: thoracoscopic radical resection of lung cancer, hidden lymph metastasis, influencing factors, predictive models, lung cancer