肺腺癌EGFR突变状态与CT纹理灰度共生矩阵的相关性研究

发布时间：2018-05-09 22:03

本文选题：肺肿瘤 + 腺癌　；参考：《中国临床医学影像杂志》2017年09期

【摘要】：目的:探讨肺腺癌表皮生长因子受体(EGFR)突变状态与其CT纹理灰度共生矩阵的相关性。方法:回顾性分析64例肺腺癌患者的临床及影像学资料,根据EGFR突变情况,分为外显子19突变组、21突变组和野生型组。使用Image J软件提取胸部CT肺窗轴位最大直径所在层面影像纹理灰度共生矩阵的五个特征值(能量、对比度、逆差矩、熵、自相关),用单因素方差检验、t检验进行统计学分析。结果:野生组的对比度均值(1 027.734)明显高于外显子19组(560.127)和外显子21组(331.987)。外显子19突变组、21突变组与野生型组在对比度、逆差距、相关性之间的差异均具有统计学意义(P0.05)。外显子19、21突变的对比度之间的差异有统计学意义(P=0.007)。实性结节占EGFR突变型患者57.8%,占野生型患者的47.3%,年龄、毛刺征、分叶征、空气支气管征等与EGFR突变无明显相关性。结论:使用灰度共生矩阵提取肺腺癌CT图像的纹理特征,对比度、相关性、逆差矩数值来描述EGFR突变的肺腺癌有较理想的结果,为预测EGFR突变提供定量分析的参数,并可能作为定量影像生物学标志物,建立肺癌影像学与基因突变之间的联系。
[Abstract]:Objective: to investigate the correlation between the mutation of epidermal growth factor receptor (EGFR) and CT texture gray level co-occurrence matrix in lung adenocarcinoma. Methods: the clinical and imaging data of 64 patients with lung adenocarcinoma were analyzed retrospectively. According to the mutation of EGFR, they were divided into exon 19 mutation group and wild type group. Image J software was used to extract the five eigenvalues (energy, contrast, deficit moment, entropy, autocorrelation) of grayscale co-occurrence matrix in the plane of the maximum diameter of lung window in chest CT, and the statistical analysis was carried out with the one-factor variance test. Results: the contrast mean value of wild group was significantly higher than that of exon 19 group (560.127) and exon 21 group (331.987). There were significant differences in contrast, inverse gap and correlation between exon 19 mutation group and wild type group (P 0.05). There was significant difference in contrast between exon 19M21 mutation and P0. 007. Solid nodules accounted for 57.8% of EGFR mutation and 47.3% of wild-type patients. Age, burr sign, lobulation sign, air bronchi sign and so on had no significant correlation with EGFR mutation. Conclusion: using gray level co-occurrence matrix to extract the texture features, contrast, correlation and deficit moment of lung adenocarcinoma CT images is an ideal method to describe lung adenocarcinoma with EGFR mutation, and to provide quantitative analysis parameters for predicting EGFR mutation. It may be used as a quantitative imaging biomarker to establish the relationship between lung cancer imaging and gene mutation.
【作者单位】：大连医科大学附属第一医院;
【分类号】：R730.44;R734.2

【相似文献】