Objective To explore the diagnostic value of CT imaging features in small cell lung cancer(SCLC).Method A total of 26 patients with SCLC were selected and included in the SCLC group.The propensity matching method was used to match gender and age in a 1∶2 ratio,a total of 52 non-small cell lung cancer(NSCLC)patients were selected and included in the NSCLC group.The CT imaging features were compared between the two groups,and incor-porate CT imaging features with statistically differences were included into multivariate Logistic regression analysis,and the influence factors of SCLC was analyzed.The receiver operating characteristic(ROC)curve was drawn,the area under the curve(AUC)was calculated,and the diagnostic value of CT imaging features for SCLC was evaluated.Result There were statistically significant differences in the classification,lesion morphology,bronchial occlusion and/or atelec-tasis status,hilar and mediastinal lymph node status,and vascular involvement between the two groups(P＜0.05).Multi-variate Logistic regression analysis showed that peripheral lung cancer and bronchial occlusion and/or atelectasis were in-dependent protective factors for SCLC(P＜0.05),while irregular tumor morphology,hilar mediastinal lymphadenopathy,and vascular involvement were independent risk factors for SCLC(P＜0.05).The ROC curve showed that morphology and vascular involvement have moderate diagnostic value for SCLC,while hilar mediastinal lymphadenopathy,bronchial occlusion and/or atelectasis status,and classification have lower diagnostic value for SCLC,and the AUC of the com-bined model in diagnosing SCLC was 0.989(95％CI:0.893-0.997),which had high diagnostic value,and higher than that of individual detection of each indicator(P＜0.05).Conclusion CT examination shows that central type,irregular lesion morphology,presence of hilar mediastinal lymphadenopathy,and vascular involvement,any one of these indicators can in-dicate SCLC,and the combined examination of these indicators has high diagnostic value for SCLC.
small cell lung cancernon-small cell lung cancerCT imaging featurediagnostic value
万方数据
维普
