摘要
目的采用受试者工作特征(ROC)曲线分析方法,对比分析能谱CT描述组织结构变化参数与肺癌病理类型的关系,为建立无创性能谱CT评估病理类型提供科学依据。方法回顾性分析山东省肿瘤医院2019-06-01-2020-12-31治疗前行能谱CT增强扫描的72例初诊肺癌患者病例资料,按照病理类型分为腺癌组35例,鳞癌组25例,小细胞癌组12例。在能谱分析软件中对所有病例的原发病灶进行感兴趣区的绘制选择,测量并计算平均值。测量病灶平扫有效原子序数(Zeff)和动脉期碘浓度,并绘制单能量能谱曲线,计算标准化碘浓度(NIC)及40~70 keV能谱曲线近段斜率(k)。采用单因素分析对3组的Zeff、NIC和k进行整体比较,独立样本t检验进行组间比较,对差异有统计学意义的参数进行ROC曲线分析。结果腺癌、鳞癌和小细胞癌3组的平扫参数Zeff、NIC及k整体差异均有统计学意义(F值分别为38.58、32.01和25.37,均P<0.05),两两多重组间比较显示均P<0.05,且参数值的变化趋势是腺癌组>鳞癌组>小细胞肺癌组。ROC诊断效能分析显示,Zeff在鉴别肺鳞癌和肺腺癌病理类型时具有较高的准确性(曲线下面积为0.911),灵敏度为74.3%,特异度为96.0%;k在鉴别肺鳞癌和小细胞肺癌时具有较高准确性(曲线下面积为0.913),灵敏度为92.0%,特异度为83.3%;Zeff在鉴别小细胞肺癌与肺腺癌病理类型时具有较高准确性(曲线下面积为0.968),灵敏度为85.7%,特异度为100.0%。结论能谱CT多参数定量分析在肺癌病理学类型的鉴别诊断中具有一定的价值,可以辅助诊断,提高诊断效能。
Objective To provide a scientific basis for the establishment of seminal spectral CT to assess pathological types by comparing and analyzing the parameters describing tissue structural changes by spectral CT with pathological types of lung cancer using ROC diagnostic curve analysis methods.Methods A total of 72 newly diagnosed lung cancer patients who underwent spectral CT enhanced scan before anti-tumor treatment from June 1,2019 to December 31,2019 in Shandong Cancer Hospital were retrospectively analyzed and divided into adenocarcinoma(ADC)group(n=35),squamous cell carcinoma(SCC)group(n=25),and small-cell lung cancer(SCLC)group(n=12)according to pathology.The region of interest(ROI)was drawn and selected for the primary lesion in all cases in the energy spectrum analysis software,measured and the average was calculated.The effective atomic number(Zeff)and iodine concentration in arterial phase were measured and the monoenergetic spectral curve was drawn.The normalized iodine concentration(NIC)and the slope k of the near segment of the spectral curve from 40 to 70 keV were calculated.Zeff,NIC and kof the three groups were compared and analyzed by SPSS 25.0 analysis software.One-way analysis of variance was used for overall comparison.Independent sample t-test was used for comparison.P<0.05 was considered statistically significant.Receiver operating characteristic(ROC)curve analysis was performed for statistically significant parameters.Results There were significant differences in Zeff,NIC and kamong the three groups in ADC group,SCC group and SCLC group(Fvalues were 38.58,32.01 and 25.37,all P<0.05).The comparison between the two groups showed that P<0.05,and the change trend of 3 parameter values was ADC group>SCC group>SCLC group.ROC diagnostic efficacy analysis showed that Zeff had high accuracy(AUC=0.911)with a sensitivity of 74.3%and a specificity of96.0%in differentiating ADC from SCC pathological types;khad high accuracy(AUC=0.913)with a sensitivity of 92.0%and a specificity of 83.3%in differentiating SCC from SCLC;and Zeff had high accuracy diagnostic value(AUC=0.968)with a sensitivity of 85.7%and a specificity of 100.0%in differentiating SCLC from ADC pathological types.Conclusion Energy spectral CT multi-parameter quantitative analysis has a certain value in the differential diagnosis of pathological types of lung squamous cell carcinoma,adenocarcinoma and small cell lung cancer.
作者
程子珊
李圣磊
李文武
崔永春
CHENG Zi-shan;LI Sheng-lei;LI Wen-wu;CUI Yong-chun(Department of Postgraduates,Shandong First Medical University,Shandong Academy of Medical Sciences,Jinan 250117,China;Shandong Cancer Hospital and Institute,Shandong First Medical University and Shandong Academy of Medical Sciences,Jinan 250117,China)
出处
《中华肿瘤防治杂志》
CAS
北大核心
2022年第1期59-65,共7页
Chinese Journal of Cancer Prevention and Treatment
关键词
肺癌
能谱成像
病理类型
诊断
lung cancer
energy spectrum imaging
pathological type
diagnosis
作者简介
第一作者:程子珊,女,山东济南人,硕士,主要从事影像诊断的研究工作。E-mail:chengzs@yeah.net;第一作者:李圣磊,男,山东肥城人,硕士,助理研究员,主要从事研究生教育管理及胸外科临床诊疗的研究工作。E-mail:lishenglei@sdfmu.edu.cn;通讯作者:李文武,男,山东济南人,研究员,硕士生导师,主要从事肿瘤影像的研究工作。E-mail:lwwzlm@outlook.com;通讯作者:崔永春,男,山东潍坊人,硕士,副研究员,硕士生导师,主要从事肿瘤流行病及卫生统计学的研究工作。E-mail:cyc0001@126.com
