甲状腺意外结节增强CT的纹理特征分析对良恶性鉴别的价值研究-东南大学学报医学版

甲状腺意外结节增强CT的纹理特征分析对良恶性鉴别的价值研究

单位：1. 东南大学医学院, 江苏南京 210009;
2. 东南大学附属中大医院医学影像科, 江苏南京 210009;
3. 南京医科大学附属无锡人民医院影像科, 江苏无锡 214023

分类号：R736.1;R445.3

出版年·卷·期（页码）：2016·35·第一期（112-116）

摘要：

目的:探讨基于灰度共生矩阵(GLCM)的纹理特征值测量技术在甲状腺良恶性结节鉴别中的价值。方法:回顾性分析经病理证实的具有甲状腺意外结节患者(46例,结节共61个)的增强CT资料,其中甲状腺良性结节23个,甲状腺恶性结节38个。使用Image J软件剪影病灶同一层面的平扫图像与动脉期图像后,测量结节内部的纹理特征值(对比度、熵和FD值)。结果:甲状腺恶性结节的熵为5.170±0.404,FD值为1.374±0.093,均高于甲状腺良性结节(熵为4.194±1.625,FD值为1.210±0.169),且组间差异有统计学意义(P<0.01)。而对比度在良恶性结节间测量差异无统计学意义(P>0.05)。经ROC曲线分析,判断FD最佳临界点为1.300,此时诊断的灵敏度为78.9%,特异度为65.2%;Entropy最佳临界点为5.000,其诊断的灵敏度为65.8%,特异度为60.9%。结论:甲状腺结节内部CT增强前后剪影图像的纹理特征值(熵和FD值)可以为定性诊断提供相对可靠的特征参数。

Objective:To investigate the value of CT texture analysis based on gray level co-occurrence matrix(GLCM) for differentiating malignant from benign thyroid nodule. Methods:The retrospective study was based on 46 cases of the incidental thyroid nodule (61 lesions) which were collected from June 2012 to May 2015.The final diagnosis was confirmed by surgery and pathology. 23 lesions were benign nodule, and 38 lesions were malignant nodule. We analyzed the CT images using Image J software as follow:firstly,pre-processing of CT images(subtraction of arterial phase and non-enhanced phase in the same slice of lesion);then, measured the value of texture parameters(entropy, contrast and FD) based on fractal theory and GLCM texture plugin. Results:Mean entropy and FD for the malignant nodules were significantly higher when compared with the benign nodules. But, mean contrast was not significantly different between malignant and benign nodules. Utilized receiver operator characteristic curve(ROC curve) in SPSS to select critical value of optimum sensitivity and specificity. When critical value of FD was 1.300,sensitivity was 78.9%,specificity was 65.2%; critical value of entropy was 5.000,sensitivity was 65.8%,specificity was 60.9%. Conclusion:Texture analysis of CT images of incidental thyroid nodule holds promise to differentiate between malignant and benign lesions.

参考文献：

[1] 赵炎斌,刘方舟,张园,等.超声弹性成像及CT检查对术前甲状腺肿物性质判断的相关性研究[J].东南大学学报:医学版,2014,3(3):296-299.
[2] TIXIER F,HATT M,le REST C C,et al.Reproducibility of tumor uptake heterogeneity characterization through textural feature analysis in 18F-FDG PET[J].J Nucl Med, 2012, 53(5):693-700.
[3] OLIVERIRA M S D,BETTING L E,MORY S B,et al.Texture analysis of magnetic resonance images of patients with juvenile myoclonic epilepsy[J].E&B, 2013,27(1):22-28.
[4] RAO S X,LAMBREGTS D M,SCHNERR R S,et al.Whole-liver CT texture analysis in colorectal cancer:does the presence of liver metastases affect the texture of the remaining liver?[J].UEG,2014,2(6):530-538.
[5] GANESHAN B,ABALEKEA S,YOUNG R,et al.Texture analysis of non-small cell lung cancer on unenhanced computed tomography:initial evidence for a relationship with tumour glucose metabolism and stage[J].Cancer Imaging,2010(10):137-143.
[6] 汪家旺,于立燕,王德杭,等.肺癌分形维数特征的研究[J].中国医学物理学杂志,2005,22(1):393-395.
[7] GROVEL O,BERGLUND A E,SCHAB M B.Quantitative computed tomographic descriptors associate tumor shape complexity and intra tumor heterogeneity with prognosis in lung adenocarcinoma[J].Pone,2015,10(3):e0118261.
[8] RANGAYYAN R M,NGUYEN T M.Fractal analysis of contours of breast masses in mammograms[J].J Digit Imaging, 2007,20(3):223-237.
[9] 邵元智,刘立志,王风华,等.不同生长方式肿瘤的CT/MRI影像观察及数据分析方法[J].中华医学杂志,2008,88(21):1503-1506.
[10] 张博,苏俊平,白杨,等.甲状腺结节合并微钙化与甲状腺癌关系及其危险因素分析[J].现代医学,2015,43(2):214-217.
[11] 李恒国,卢绍辉,梁久平,等.甲状腺结节的CT灌注评价[J].中华放射学杂志,2011,45(9):831-834.
[12] GANESHAN B,MILES K A,YOUNG R.Texture analysis in non-contrast enhanced CT:impact of malignancy on texture in apparently disease-free areas of the liver[J].Eur J Radiol,2009,70:101-110.
[13] BROOKS F J, GRIGSBY P W.The effect of small tumor volumes on studies of intratumoral heterogeneity of tracer uptake[J].J Nucl Med,2014,55(1):37-42.

服务与反馈：

【文章下载】【发表评论】【查看评论】【加入收藏】

提示：您还未登录，请登录！点此登录