肺部纯磨玻璃密度结节浸润前与浸润性病变MSCT诊断价值研究

发布时间：2018-02-23 23:00

本文关键词： 肺腺癌纯磨玻璃密度结节计算机断层成像均匀度　出处：《吉林大学》2017年硕士论文　论文类型：学位论文

【摘要】：目的:探讨肺部纯磨玻璃密度结节(pure groung-glass opacity nodule,p GGN)的CT影像特征及临床特点对于浸润前病变(preinvasive lesion,PIL)[包括不典型腺瘤样增生(atypical adenomatous hyperplasia,AAH)、原位腺癌(adenoeareinoma in situ,AIS)]及浸润性病变[包括微浸润腺癌(minimally invasive adenocarcinoma,MIA)、浸润性肺腺癌(invasive lung adenocarcinomas,ILA)]的诊断价值,尝试定量分析p GGN均匀度对浸润前与浸润性病变的鉴别诊断价值。材料与方法:回顾性分析2014年9月至2016年12月期间被吉林大学第一医院收治并行胸部CT(computed tomography)检查,经术后病理证实的80例p GGN的CT影像特征及临床特点,包括病变的位置、大小、密度、边缘(光滑、分叶征、毛刺征)、瘤-肺界面(清晰、不清晰)、内部征象(空气支气管征、空泡征、囊腔)、均匀度、胸膜凹陷征、患者性别、年龄、吸烟史。选取p GGN内任意三点感兴趣区(region of interest,ROI),测量平均CT值及标准差,需避开血管及支气管,取三个ROI变异系数[(标准差SD/平均值Mean)×100%),C·V]平均值的相对值代表病灶均匀度。病灶大小、均匀度、密度及患者年龄在病理分组之间的比较采用独立样本t检验;患者性别、病变部位及CT影像特征的比较采用卡方检验;以P0.05为差异具有统计学意义,并通过受试者工作曲线(receive operating characteristic,ROC)分析浸润前与浸润性病变大小及均匀度的最佳临界值。结果:80例p GGNs中浸润前病变16例(AAH 6例,AIS 10例);浸润性病变64例(MIA 29例,ILA 35例)。在浸润前病变与浸润性病变之间,患者性别、年龄、病灶部位及空泡征的差异无统计学意义。病灶大小、均匀度、密度、分叶征、毛刺征、瘤-肺界面、空气支气管征及胸膜凹陷征在鉴别浸润前与浸润性病变之间差异具有统计学意义(各组P值分别为0.001、0.001、0.01、0.033、0.034、0.029、0.041、0.042,P0.05)。ROC曲线显示以直径10.5mm为鉴别浸润前病变与浸润性病变的临界值时,敏感度为81.3%,特异性为71.9%,ROC曲线下面积(AUC)为0.770;以0.085作为鉴别浸润前与浸润性病变密度均匀度的最佳临界值,敏感度为81.3%,特异性为75.0%,AUC为0.785。结论:病灶的大小、CT影像特征(均匀度、密度、分叶征、毛刺征、瘤-肺界面、空气支气管征、胸膜凹陷征)对p GGN浸润前与浸润性病变具有鉴别诊断价值;定量分析p GGN均匀度,为评价病灶均匀度提供一种新的客观分析方法。
[Abstract]:Objective: to investigate the CT features and clinical features of pure groung-glass opacity noduleus (GGNN) for preinvasive lesions (including atypical adenomatous hyperplasia, adenoeareinoma in situ) and invasive lesions [including microinvasive lesions]. The diagnostic value of minimally invasive invasive adenocarcinoma A and invasive lung adenocarcinomassus (ILA). Materials and methods: from September 2014 to December 2016, we retrospectively analyzed the value of p GGN evenness in the differential diagnosis of preinvasive and invasive lesions by CT(computed CT(computed examination in the first Hospital of Jilin University, which was admitted to the first Hospital of Jilin University from September 2014 to December 2016. Ct features and clinical features of 80 cases of p GGN confirmed by postoperative pathology, including location, size, density, margin (smooth, lobular sign, burr sign, tumor-lung interface (clear, unclear), internal sign (air bronchus sign) of the lesion, were analyzed. Vacuole sign, cystic cavity, uniformity, pleural indentation, patient sex, age, smoking history. Select any three regions of interest in p GGN to measure average CT value and standard deviation, avoid blood vessels and bronchi, The relative values of the mean values of the three ROI coefficients of variation [(standard deviation mean / mean) 脳 100C 路V] were taken to represent the lesion evenness. The size, evenness, density and age of the lesions were compared between pathological groups by independent sample t test. The location of lesion and CT imaging features were compared by chi-square test, with P0.05 as the difference was statistically significant, The optimal critical value of the size and evenness of the infiltrating lesions before infiltration and with the invasive lesions was analyzed by the operating curve of the subjects. Results in the 80 cases of p GGNs, 16 cases had preinvasive lesions, 6 cases had AIS, and 64 cases had infiltrative lesions with MIA in 29 cases. Between preinvasive lesions and invasive lesions, There was no significant difference in sex, age, location of lesion and vacuole sign. Lesion size, uniformity, density, lobulation, burr sign, tumor-lung interface, There were significant differences between air bronchus sign and pleural depression sign in differentiating preinvasive lesions from invasive lesions (P = 0.001 / 0. 001 / 0. 01 / 0. 031 / 0. 033 / 0. 034 / 0. 029 / 0. 041 / 0. 042) P 0.05 / ROC curve indicating that the diameter of 10. 5mm was taken as the critical value for differentiating preinvasive lesions from invasive lesions. The sensitivity is 81.3, the specificity is 71.9 and the area under the ROC curve is 0.770; the best critical value for distinguishing the density uniformity between the pre-invasive and the invasive lesions is 0.085, the sensitivity is 81.3 and the specificity is 75.0 and the AUC is 0.785.Conclusion: the size of the lesion is uniform. Density, lobulation sign, burr sign, tumor-lung interface, air bronchus sign, pleural indentation sign) are valuable for differential diagnosis of preinvasive and invasive lesions of p GGN, quantitative analysis of p GGN uniformity, It provides a new objective analysis method for evaluating the evenness of lesions.
【学位授予单位】：吉林大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：R734.2;R730.44

【相似文献】