摘要
目的评估CT图像纹理特征对≥6 mm纯磨玻璃密度肺腺癌中浸润性病变的鉴别价值。方法回顾性分析2013年9月至2015年10月期间上海长征医院所有高分辨率CT图像表现为≥6 mm肺部纯磨玻璃结节(pGGN)且病理结果明确的病例,收集其基线临床资料。共有来自91例患者的91个pGGN纳入研究,依照病理结果分为浸润前组(n=39)和浸润性组(n=52)。对每个pGGN沿分界进行半自动分割并提取该区域的图像纹理特征参数,分析比较两组患者的临床数据和图像纹理特征参数。采用Logistic回归分析浸润性病变的独立鉴别指标,接受者操作特性(ROC)曲线分析各独立指标及回归模型的鉴别诊断效能。结果平均CT值、最大CT值、最大有效长径、表面积、体积、质量及逆差距在两组间具有统计学差异(P<0.05)。Logistic回归结果证明,逆差距(OR=0.559)、最大有效长径(OR=1.305)及平均CT值(OR=1.009)为浸润性病变的独立鉴别指标(P<0.05)。使用Logistic回归模型鉴别浸润性病变的ROC曲线下面积(AUC)为0.809,鉴别效能优于单独使用逆差距(AUC=0.672)、平均CT值(AUC=0.660)及最大有效长径(AUC=0.704)。结论图像纹理参数分析能够较为准确地鉴别≥6 mm pGGN中的浸润性病变,逆差距、平均CT值及最大有效长径为独立鉴别指标。
Objective To evaluate the performance of computerized texture analysis in the diagnosis of invasive adenocarcinoma manifested as pure ground-glass nodules 6 mm or larger.Methods All the patients with pathologically confirmed pure ground-glass nodules(pGGNs)6 mm or larger between September 2013 and October 2015 were retrospectively reviewed,and the baseline clinical data was collected.A total of 91 pGGNs from 91 patients were included.Patients were divided into the preinvasive group(n=39)and invasive group(n=52)according to the results of pathology.All PGGNs were semiautomatically segmented on images and computerized texture features were extracted.Patients’clinical data and computerized texture features were compared between the two groups.Binary Logistic regression analysis was used to investigate the differentiating factors of invasive lesions.The discriminating performance of the texture features was evaluated by ROC curve analysis.Results The mean CT value,maximum CT value,maximum diameter,volume,surface area,mass and inverse difference moment(IDM)were significantly different between the two groups(P<0.05).Binary Logistic regression analysis revealed that the IDM(adjusted OR=0.559),maximum diameter(adjusted OR=1.305)and mean CT value(adjusted OR=1.009)were independent predictors of invasive lesions.A combination of these features showed better differentiatingperformance(AUC=0.809)than the IDM(AUC=0.672),mean CT value(AUC=0.660),or maximum diameter(AUC=0.704)used alone.Conclusion Computerized texture analysis can differentiate lung invasive adenocarcinoma manifested as pGGNs with a diameter≥6 mm.The IDM,mean CT value and maximum diameter are independent differentiators.
作者
李西
肖湘生
董伟华
LI Xi;XIAO Xiang sheng;DONG Wei hua(Department of Interventional Radiology,Changzheng Hospital,Naval Medical University(Second Military Medical Uni versity),Shanghai 200003,China)
出处
《军事医学》
CAS
北大核心
2019年第11期871-874,共4页
Military Medical Sciences
