Evaluation value of preoperative mediastinal lymph node metastasis in non-small cell lung cancer using enhanced CT combined with texture analysis

LIU Yang; LI Xinzhen; XIA Liang; CHANG Xiaoyue

doi:10.20258/j.cnki.1006-9011.2025.03.012

Abstract：Objective To explore the evaluation value of enhanced CT scanning combined with texture analysis for preopera-tive mediastinal lymph node metastasis in non small-cell lung cancer(NSCLC).Methods A total of 110 NSCLC patients con-firmed by biopsy pathology in our hospital were selected.All patients underwent enhanced CT scanning and texture analysis ex-amination.After completing the CT examination,lung cancer radical surgery was performed.The mediastinal lymph node metas-tasis was evaluated according to the results of surgical pathology.The results of pathological specimens collected by surgery were used as the gold standard to evaluate the value of enhanced CT scan combined with texture analysis in the evaluation of mediasti-nal lymph node metastasis of NSCLC.Results A total of 110 NSCLC patients underwent radical surgery for lung cancer,and mediastinal lymph node dissection or sampling surgery was performed based on intraoperative exploration.Pathological diagnosis showed positive lymph nodes in 35 cases,with a mediastinal lymph node metastasis rate of about 31.82%.The evaluation of me-diastinal lymph node metastasis in NSCLC by enhanced CT and CT texture analysis showed moderate consistency with pathologi-cal results(Kappa=0.439,0.463,P<0.05).The use of enhanced CT combined with texture analysis to evaluate mediastinal lymph node metastasis in NSCLC showed a high degree of consistency with pathological results(Kappa=0.734,P<0.05),and the accuracy and specificity of the combined evaluation were higher than those of enhanced CT and texture analysis alone(P<0.05).Conclusion The evaluation of mediastinal lymph node metastasis in NSCLC using enhanced CT scanning and texture analysis is consistent with pathological results,and the consistency is strongest when evaluated in combination,with better value compared to their individual application or evaluation.