CT三维重建在早期微小肺结节鉴别诊断中的应用  

Application of CT 3D Reconstruction in Differential Diagnosis of Minute Pulmonary Nodules

作　　者：王廷婷 徐莹 林保光 吴彤 姜曈 吕东阳 WANG Ting-ting;XU Ying;LIN Bao-guang;WU Tong;JIANG Tong;LV Dong-yang(Department of Radiotherapy,General Hospital of Northern Theatre Command,Shenyang 110000,Liaoning Province,China)

机构地区：[1]中国人民解放军北部战区总医院放射治疗科,辽宁沈阳110000

出　　处：《中国CT和MRI杂志》2025年第3期75-78,共4页Chinese Journal of CT and MRI

基　　金：辽宁省科学技术计划项目(2020JH/10300172)。

摘　　要：目的探究电子计算机断层扫描(CT)三维重建在微小肺结节鉴别诊断中的应用价值。方法收集癌症影像在线数据集(TCIA)2018年11月至2020年3月的171例微小肺结节患者为研究对象,所有患者均接受胸部CT扫描、CT三维重建检查,分析良恶性肺结节CT三维重建影像学特征,并分析CT三维重建对微小肺结节的诊断价值。结果病理诊断结果显示171例患者共计171个微小肺结节,良性74个,恶性97个;良性结节与恶性结节患者在空泡征、血管集束征、分叶征、毛刺征、胸膜牵拉征差异均有统计学意义(P<0.05),在形状、实质成分、钙化、棘突征、空气支气管征、空洞特征差异均无统计学意义(P>0.05);Logistic回归分析显示,空泡征、血管集束征、分叶征、毛刺征、胸膜牵拉征是良恶性结节的危险因素(P<0.05);受试者工作曲线(ROC)结果显示,单项指标诊断微小肺结节良恶性时,毛刺征诊断的AUC最高为0.869,敏感度为84.54%,特异度为89.19%,联合诊断微小肺结节时,空泡征、血管集束征、分叶征、毛刺征、胸膜牵拉征联合诊断的AUC最高为0.968,敏感度为94.85%,特异度为89.19%。结论CT三维重建诊断早期微小肺结节具有良好的诊断效能。Objective To investigate the application value of 3D reconstruction of computed tomography(CT)in differential diagnosis of minute pulmonary nodules.Methods From November 2018 to March 2020,171 patients with minute pulmonary nodules from The Cancer Imaging Archive(TCIA)were selected as the research subjects.All of them were examined with chest CT and CT 3D reconstruction.CT 3D reconstruction imaging features of benign and malignant nodules were analyzed.The diagnostic value of CT 3D reconstruction for minute pulmonary nodules was evaluated.Results Pathological results showed that the 171 patients had a total of 171 minute pulmonary nodules,including 74 benign ones and 97 malignant ones.There were statistically significant differences in vacuolar sign,vascular convergence sign,lobulation sign,spicule sign,and pleural traction sign between patients with benign and malignant nodules(P<0.05).However,there were no statistically significant differences in shape,composition of the parenchyma,calcification,spinous process sign,air bronchogram sign,and cavity features(P>0.05).Logistic regression analysis showed that vacuolar sign,vascular convergence sign,lobulation sign,spicule sign,and pleural traction sign were risk factors for benign and malignant nodules(P<0.05).The receiver operating characteristic(ROC)curve analysis results showed that when a single indicator was used to diagnose benign or malignant minute pulmonary nodules,the AUC of spicule sign was the largest,which was 0.869.Its sensitivity and specificity were 84.54%and 89.19%.For joint diagnosis,the AUC of combination of vacuolar sign,vascular convergence sign,lobulation sign,spicule sign,and pleural traction sign was the largest,which was 0.968.The sensitivity and specificity were 94.85%and 89.19%.Conclusion CT 3D reconstruction has good diagnostic efficacy for early small lung cancer.

关 键 词：微小肺结节 早期 电子计算机断层扫描 三维重建 诊断 

分 类 号：R563[医药卫生—呼吸系统]