平扫T2脂肪抑制序列图像纹理可提高诊断乳腺良恶性结节的准确率

陈文静; 牟伟; 张文馨; 徐蕊; 燕桂新; 梁颖

doi:10.12122/j.issn.1674-4500.2019.04.07

平扫T2脂肪抑制序列图像纹理可提高诊断乳腺良恶性结节的准确率

doi: 10.12122/j.issn.1674-4500.2019.04.07

陈文静^1,,
牟伟²,
张文馨¹,
徐蕊¹,
燕桂新¹,
梁颖^3, ,

1.
五家渠市兵团第六师医院影像科，新疆五家渠 831300
2.
美国Moffitt癌症研究中心，佛罗里达坦帕 33612
3.
国家癌症中心//中国医学科学院北京协和医学院肿瘤医院//深圳医院PET-CT中心，广东深圳 518116

基金项目: 新疆兵团卫生局科研项目（1520）

详细信息

作者简介:
陈文静，硕士，副主任医师，E-mail：wen-jing333@163.com

通讯作者:
梁颖，博士，主任医师，E-mail：langy_2000@sina.com

计量
- 文章访问数: 1182
- HTML全文浏览量: 448
- PDF下载量: 30
- 被引次数: 0
出版历程
- 收稿日期: 2019-08-15
- 刊出日期: 2019-12-01

Value of T2 fat inhibition sequence image texture analysis in diagnosis of benign and malignant breast nodules

1.
Department of Radiology, the Sixth Division Hospital of the Xinjiang Production and Construction Corps, Wujiaqu 831300, China
2.
H. Lee Moffitt Cancer Center & Research Institute, Tampa 33612, USA
3.
National Cancer Center/Chinese Academy of Medical Sciences & Peking Union Medical College, PET-CT Center, Shenzhen Hospital, Shenzhen 518116, China

摘要

摘要: 目的探讨平扫T2脂肪抑制序列（FS T2WI）图像纹理分析鉴别诊断乳腺良恶性结节的价值。方法回顾性分析经手术病理证实的60例患者共61个乳腺结节FS T2WI图像。绘制纹理参数鉴别诊断良恶性乳腺结节的ROC曲线，并与病理结果进行对照。结果60例患者的61个乳腺结节，FS T2WI纹理参数灰度区域矩阵重点运行高灰度级判断乳腺结节良恶性的AUC值（0.701）最大且诊断准确率高，其诊断恶性乳腺结节的敏感度为65.52%（19/29），特异度为71.88%(23/32)，误判率为31.15%（19/61）。FS T2WI图像诊断乳腺恶性结节敏感度为71.43%（20/28）、特异度为63.64%（21/33）、误判率为32.79%（20/61）；两者联合应用敏感度为64.29%(18/28)，特异度78.79%（26/33），误判率为27.87%（15/61）,与单独T2比较，差异有统计学意义（χ²=72.255，P=0.000）。结论平扫FS T2WI序列结合纹理分析能够提高诊断乳腺良恶性结节的特异度，降低误判率，提高乳腺平扫诊断的准确度。
- 磁共振成像 /
- 脂肪抑制 /
- 纹理分析 /
- 乳腺肿瘤
Abstract: ObjectiveTo investigate the value of image texture analysis of T2 Fat-suppression sequence (FS T2WI) in differential diagnosis of benign and malignant breast nodules.MethodsA total of 61 images of breast nodules FS T2WI were retrospectively analyzed in 60 patients with surgical pathology. The ROC curves for differential diagnosis of benign and malignant breast nodules were drawn and compared with the pathological results.Results61 breast nodules in 60 patients, The FS T2WI texture parameter gray region matrix focuses on the operation of high gray level to determine the AUC value (0.701) of benign and malignant breast nodules with high diagnostic accuracy. The sensitivity of the diagnosis of malignant breast nodules was 65.52 % (19/29), and the specificity was 71.88 % (23/32), and the misjudgment rate was 31.15 % (19/61). The diagnostic sensitivity of FS T2WI images to malignant nodules in the breast was 71.43 %(20/28), and the specificity was 63.64 % (21/33), and the miscalculation rate was 32.79 %(20/61); The combined application sensitivity of the two is 64.29 %(18/28), the specificity is 78.79 % (26/33), and the miscalculation rate is 27.87 %(15/61). Compared with individual T2, the difference is statistically significant (χ²=72.255, P=0.000).ConclusionFS T2WI sequence combined with texture analysis can improve the specificity of diagnosis of benign and malignant breast nodules, which can reduce the rate of misjudgment, and improve the accuracy of diagnosis.
- magnetic resonance imaging /
- fat inhibition /
- texture analysis /
- breast tumors

HTML全文

表 1 FS T2WI与纹理分析联合诊断与病理结果对比（n）

FS T2WI与纹理联合病理合计
阳性阴性

阳性 18 7 25
阴性 10 26 36
合计 28 33 61

下载: 导出CSV

参考文献(24)

[1]	Saslow D, Boetes C, Burke W, et al. American cancer society guidelines for breast screening with MRI as an adjunct to mammography[J]. CA Cancer J Clin, 2007, 57(2): 75-89. doi: 10.3322/canjclin.57.2.75
[2]	许玲辉, 顾雅佳. 乳腺磁共振在乳腺癌诊断治疗中的运用[J]. 中国癌症杂志, 2013, 23(8): 613-7. doi: 10.3969/j.issn.1007-3969.2013.08.009
[3]	李珂, 刘惠. 乳腺计算机辅助诊断中DCE-MRI图像特征的选择与分析[J]. 北京生物医学工程, 2012, 31(4): 343-8. doi: 10.3969/j.issn.1002-3208.2012.04.03.
[4]	Waugh SA, Purdie CA, Jordan LB, et al. Magnetic resonance imaging texture analysis classification of primary breast cancer[J]. Eur Radiol, 2016, 26(2): 322-30. doi: 10.1007/s00330-015-3845-6
[5]	Wu M, Ma J. Association between imaging characteristics and different molecular subtypes of breast cancer[J]. Acad Radiol, 2017, 24(4): 426-34. doi: 10.1016/j.acra.2016.11.012
[6]	吴佩琪, 赵可, 梁长虹, 等. 基于扩散加权成像和动态增强MRI的影像组学特征与乳腺癌分子分型的关系初探[J]. 中华放射学杂志, 2018, 52(5): 338-43. doi: 10.3760/cma.j.issn.1005-1201.2018.05.004
[7]	曹崑, 刘慧, 孙应实, 等. 常规MRI纹理分析鉴别乳腺良,恶性病变的价值初探[J]. 中华放射学杂志, 2017, 51(8): 588-91. doi: 10.3760/cma.j.issn.1005-1201.2017.08.006
[8]	黄远明, 梁立华, 陈晓东, 等. 钼靶及MRI纹理分析技术在乳腺疾病诊断中的研究进展[J]. 中国CT和MRI杂志, 2019, 17(6): 147-50. doi: 10.3969/j.issn.1672-5131.2019.06.045
[9]	Nketiah G, Elschot M, Kim E, et al. T2-weighted MRI-derived textural features reflect prostate cancer aggressiveness: preliminary results[J]. Eur Radiol, 2017, 27(7): 3050-9. doi: 10.1007/s00330-016-4663-1
[10]	Choi Y, Kim SH, Youn IK, et al. Rim sign and histogram analysis of apparent diffusion coefficient values on diffusion-weighted MRI in triple-negative breast cancer: comparison with ER-positive subtype[J]. PLoS One, 2017, 12(5): e0177903-14. doi: 10.1371/journal.pone.0177903
[11]	陈瑾, 王海屹, 叶慧义. 纹理分析在肿瘤影像学中的研究进展[J]. 中华放射学杂志, 2017, 51(8): 979-82.
[12]	宋之琰. MRI纹理分析在乳腺癌诊疗中的应用进展[J]. 影像诊断与介入放射学, 2019, 28(3): 219-24. doi: 10.3969/j.issn.1005-8001.2019.03.011
[13]	邓丹琼, 梁碧玲. 乳腺癌患者临床病理特征与磁共振成像强化率关系[J]. 中国公共卫生, 2016, 32(9): 1255-7. doi: 10.11847/zgggws2016-32-09-32
[14]	American College of Radiology. Breast imaging reporting and data system[M]. 4th Edition, Reston: Breast Imaging Atlas, 2003.
[15]	牟玮. 肿瘤PET/CT成像影像组学相关算法的研究[D]. 北京: 中国科学院大学, 2016: 42-6.
[16]	吴晓燕, 张贵祥, 李康安, 等. 增强与非增强组合序列MRI对乳腺病变筛查的比较研究[J]. 放射学实践, 2014, 29(12): 1424-8.
[17]	Uematsu T, Kasami M, Watanabe J. Is evaluation of the presence of prepectoral edema on T2-weighted with fat-suppression 3T breast MRI a simple and readily available noninvasive technique for estimation of prognosis in patients with breast cancer[J]. Breast Cancer, 2014, 21(6): 684-92. doi: 10.1007/s12282-013-0440-z
[18]	Dong Y, Feng Q, Yang W, et al. Preoperative prediction of sen tinel lymph node metastasis in breast cancer based on radiom ics of T2-weighted fat-suppression and diffusion-weighted MRI[J]. Eur Radiol, 2018, 28(6): 582-91.
[19]	Bhooshan N, Giger M, Lan L, et al. Combined use of T2-weighted MRI and T1-weighted dynamic contrast-enhanced MRI in the automated analysis of breast lesions[J]. Magn Reson Med, 2011, 66(2): 555-64. doi: 10.1002/mrm.22800
[20]	Rakoczy M, Mcgaughey D, Korenberg MJ, et al. Feature selection in computer-aided breast cancer diagnosis via dynamic contrast-enhanced magnetic resonance images[J]. J Digit Imaging, 2013, 26(2): 198-208. doi: 10.1007/s10278-012-9506-2
[21]	Wang TC, Huang YH, Huang CS, et al. Computer-aided diagnosis of breast DCE-MRI using pharmacokinetic model and 3-D morphology analysis[J]. Magn Reson Imaging, 2014, 32(3): 197-205. doi: 10.1016/j.mri.2013.12.002
[22]	Kawshima H, Kobayashi YM. Peripheralhy perintense pattern pn T2-weighted magneric resonance imaging(MRI)in breast carcinoma:correlation with early peripheral en-hancement on dynamic MRI and histopathologic findings[J]. J Magn Reson Imaging, 2010, 32(5): 1117-23. doi: 10.1002/jmri.22279
[23]	李汉森, 章强, 齐海坤, 等. 基于增强 MRI 纹理分析区分非产褥期乳腺炎与非肿块样强化病灶乳腺癌[J]. 中国医学影像学杂志, 2017, 25(3): 354-9.
[24]	Clendenen TV, Kim S, Moy L, et al. Magnetic resonance imaging(MRI)of hormone-induce breast changes in young premeno-pausal women[J]. Magn Reson Imag, 2013, 106(31): 1-9.