Please wait a minute...
J Zhejiang Univ (Med Sci)  2018, Vol. 47 Issue (4): 400-404    DOI: 10.3785/j.issn.1008-9292.2018.08.12
Diffusion-weighted imaging texture features in differentiation of malignant from benign nonpalpable breast lesions for patients with microcalcifications-only in mammography
CHEN Shujun1,2(),SHAO Guoliang2,SHAO Feng3,ZHANG Minming1,*()
1. Department of Radiology, the Second Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou 310009, China
2. Department of Radiology, Zhejiang Cancer Hospital, Hangzhou 310022, China
3. Department of Gynecologic Oncology, Zhejiang Cancer Hospital, Hangzhou 310022, China
Download: HTML( 5 )   PDF(971KB)
Export: BibTeX | EndNote (RIS)      


Objective: To evaluate the application of MR diffusion-weighted imaging(DWI) texture features in differentiation of malignant from benign nonpalpable breast lesion for patients with microcalcifications-only in mammography. Methods: The clinical and MR-DWI data of 61 patients with microcalcifications, who underwent three-dimensional positioning of breast X-ray wire from October 2012 to December 2015 in Zhejiang Cancer Hospital, were retrospectively analyzed, including 38 patients with malignant lesions and 23 patients with benign lesions. Two radiologists independently drew the regions of interest (ROI) on DWI for image segmentation, and 6 histogram features and 16 grayscale symbiosis matrix (GLCM) texture features were extracted on each ROI. The random forest algorithm was applied to select the features and built the classification model. The leave-one-out cross-validation (LOOCV) was used to validate the classifier, and the performance of the classifier was evaluated by ROC curve. Results: Six features were selected, including histogram features of mean, variance, skewness, entropy, as well as contrast (0°) and correlation (45°) in GLCM. The histogram features of mean, variance, skewness and entropy were significantly different between the benign and malignant breast lesions (all P < 0.05). The AUC of the model was 0.76, and the diagnostic accuracy, sensitivity and specificity were 77.05%, 84.21% and 65.21%, respectively. Conclusions: The texture feature analysis of DWI can improve the diagnostic accuracy of differentiating benign and malignant breast nonpalpable lesions with microcalcifications-only in mammography. Histogram features of mean, variance, skewness, entropy of DWI may be used as important imaging markers.

Key wordsBreast neoplasms/diagnosis      Breast diseases/diagnosis      Diffusion magnetic resonance imaging      Diagnosis, differential     
Received: 20 June 2018      Published: 04 December 2018
CLC:  R445.2  
Corresponding Authors: ZHANG Minming     E-mail:;
Cite this article:

CHEN Shujun,SHAO Guoliang,SHAO Feng,ZHANG Minming. Diffusion-weighted imaging texture features in differentiation of malignant from benign nonpalpable breast lesions for patients with microcalcifications-only in mammography. J Zhejiang Univ (Med Sci), 2018, 47(4): 400-404.

URL:     OR


目的: 探索使用磁共振弥散加权成像(DWI)的纹理特征鉴别良恶性单纯微钙化型乳腺隐匿性病灶的价值。方法: 回顾性分析2012年10月至2015年12月于浙江省肿瘤医院进行乳腺X射线金属丝三维定位的61例(38例恶性肿瘤、23例良性病变)单纯微钙化型乳腺隐匿性病灶患者的DWI影像。对每个感兴趣区域提取6个直方图特征和16个灰度共生矩阵纹理特征。利用随机森林算法进行特征选择并构建分类器,使用留一法进行交叉验证。采用ROC曲线评价分类器的性能。结果: 筛选获得的性能最佳的特征组合共包含6个特征,分别为直方图中的均值、方差、偏度、熵,以及灰度共生矩阵的0°方向对比度、45°方向相关性,其中,直方图中的均值、方差、偏度和熵在良性病变与恶性肿瘤组间差异有统计学意义(均P < 0.05)。由上述特征构建的分类器鉴别诊断单纯微钙化型乳腺隐匿性病灶良性病变与恶性肿瘤的AUC达到0.76,诊断准确率、敏感度、特异度分别为77.05%、84.21%、65.21%。结论: DWI的纹理特征分析能在一定程度上提高良恶性单纯微钙化型乳腺隐匿性病灶诊断的准确性,直方图中的均值、方差、偏度和熵是极具潜力的影像学标志物。

关键词: 乳腺肿瘤/诊断,  乳腺疾病/诊断,  磁共振成像, 弥散,  诊断, 鉴别 
($\bar x \pm s$)
组别 n直方图特征 灰度共生矩阵特征
均值 方差 偏度 0°方向对比度 45°方向相关性
良性病变组 23 1.70±0.35 44 514±18 401 -0.82±0.44 6.7±0.7 -458 518±282 087 0.000 16±0.000 12
恶性肿瘤组 38 1.19±0.25 50 161±18 401 -0.48±0.50 6.3±0.5 -358 601±274 809 0.000 11±0.000 13
t 3.77 2.02 2.31 2.31 1.89 1.73
P <0.01 <0.05 <0.05 <0.05 >0.05 >0.05
Tab 1 Features of classifier and the statistical differences between benign and malignant groups
Fig 1 The ROC curve of the classifier in distinguishing of benign and malignant tumors
[1]   BRAY F , FERLAY J , SOERJOMATARAM I et al. Global cancer statistics 2018:GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries[J]. Cancer J Clin, 2018, 0 1- 31
[2]   周易冬 . 不可触及的乳腺病变的诊断[J]. 中国实用外科杂志, 1997, 17 (4): 243- 245
ZHOU Yidong . Diagnose of nonpalpable breast lesion[J]. Chinese Journal of Practical Surgery, 1997, 17 (4): 243- 245
doi: 10.3321/j.issn:1005-2208.1997.04.026
[3]   Screening and Immunisations Team, Health and Social Care Information Centre. Breast screening programme, England statistics for 2013-14[R/OL]. (2015-02-18)[2015-02-19].
[4]   Netherlands Centre for Population Screening. Breast cancer screening factsheet[R/OL]. (2015-11-11)[2018-05-12].
[5]   BASSETT L W . Mammographic analysis of calcifications[J]. Radiol Clin North Am, 1992, 30 (1): 93- 105
[6]   KIM S Y , KIM H Y , KIM E K et al. Evaluation of malignancy risk stratification of microcalcifications detected on mammography:a study based on the 5th edition of BI-RADS[J]. Ann Surg Oncol, 2015, 22 (9): 2895- 2901
doi: 10.1245/s10434-014-4362-6
[7]   陈淑君, 戴罡, 陈杰 et al. 数字乳腺X线金属丝三维立体定位技术在乳腺隐匿性病灶中的临床应用价值[J]. 中国现代医生, 2015, 53 (6): 74- 77
CHEN Shujun , DAI Gang , CHEN Jie et al. The clinical value of applicate three-dimensional digital breast X-ray wire positioning technology in nonpalpable breast lesion[J]. China Modern Doctor, 2015, 53 (6): 74- 77
[8]   徐晓洲, 宋颖, 张柏林 et al. X线立体定位真空辅助活组织检查对乳腺微小钙化灶的诊断价值[J]. 肿瘤研究与临床, 2015, 27 (4): 243- 245
XU Xiaozhou , SONG Ying , ZHANG Bolin et al. Diagnostic value of X-ray stereotactic vacuum-assisted biopsy for breast micro-calcification[J]. Cancer Research and Clinic, 2015, 27 (4): 243- 245
doi: 10.3760/cma.j.issn.1006-9801.2015.04.006
[9]   EI K R H , JACOBS M A , MEZBAN S D et al. Diffusion-weighted imaging improves the diagnostic accuracy of conventional 3.0-T breast MR imaging[J]. Radiology, 2010, 256 (1): 64- 73
doi: 10.1148/radiol.10091367
[10]   BICKELHAUPT S , PAECH D , KICKINGEREDER P et al. Prediction of malignancy by a radiomic signature from contrast agent-free diffusion MRI in suspicious breast lesions found on screening mammography[J]. J Magn Reson Imaging, 2017, 46 (2): 604- 616
doi: 10.1002/jmri.v46.2
[11]   BICKELHAUPT S, STEUDLE F, PAECH D, et al. On a fractional order calculus model in diffusion weighted breast imaging to differentiate between malignant and benign breast lesions detected on X-ray screening mammography[J/OL]. PLoS One, 2017, 12(4): e0176077.
[12]   KOH D M , COLLINS D J . Diffusion-weighted MRI in the body:applications and challenges in oncology[J]. AJR Am J Roentgenol, 2007, 188 (6): 1622- 1635
doi: 10.2214/AJR.06.1403
[13]   CAI H, PENG Y, OU C, et al. Diagnosis of breast masses from dynamic contrast-enhanced and diffusion-weighted MR: a machine learning approach[J/OL]. PLoS One, 2014, 9(1): e87387.
[14]   PAREKH V , JACOBS M A . Radiomics:a new application from established techniques[J]. Expert Rev Precis Med Drug Dev, 2016, 1 (2): 207- 226
doi: 10.1080/23808993.2016.1164013
[15]   FAN M , HE T , ZHANG P et al. Heterogeneity of diffusion-weighted imaging in tumours and the surrounding stroma for prediction of Ki-67 proliferation status in breast cancer[J]. Sci Rep, 2017, 7 (1): 2875
[16]   GRANITTO P M , FURLANELLO C , BIASIOLI F et al. Recursive feature elimination with random forest for PTR-MS analysis of agroindustrial products[J]. Chemometr Intell Lab Syst, 2006, 83 (2): 83- 90
doi: 10.1016/j.chemolab.2006.01.007
[17]   LIAW A , WIENER M . Classification and regression by randomForest[J]. R News, 2002, 5 (1): 983- 999
[18]   ADAMOPOULOS P. A study of cross-validation and bootstrap for accuracy estimation and model selection[C]//IJCAI'95 Proceedings of the 14th international joint conference on Artificial intelligence-Volume 2. San Francisco: Morgan Kaufmann Publishers Inc, 1995: 1137-1143.
[19]   BAKER R , ROGERS K D , SHEPHERD N et al. New relationships between breast microcalcifications and cancer[J]. Br J Cancer, 2010, 103 (7): 1034- 1039
doi: 10.1038/sj.bjc.6605873
[20]   YAO D , YANG J , ZHAN X et al. A novel random forests-based feature selection method for microarray expression data analysis[J]. Int J Data Min Bioinform, 2015, 13 (1): 84- 101
doi: 10.1504/IJDMB.2015.070852
[1] LI Aijing, PAN Yuning, CHEN Bin, XIA Jianbi, GAN Fang, JIN Yinhua, ZHENG Jianjun. Association of parameters in dynamic contrast-enhanced MRI using reference region model with prognostic factors and molecular subtypes of breast cancer[J]. J Zhejiang Univ (Med Sci), 2017, 46(5): 505-510.
[2] SONG Rui-rui, YU Xin-feng, YERFAN·Jiaerken, SUN Jian-zhong, MAO Ying-ying, GUO Yang, CHEN Zhi-cai, ZHANG Min-ming. Cerebral lesions of DWI hyperintensity in patients with subacute stroke assessed by intravoxel incoherent motion technique[J]. J Zhejiang Univ (Med Sci), 2015, 44(6): 632-637,644.
[3] CAO Pei-long,et al. Expressions of FOXC1 and MMP-7 in molecular subtypes of breast cancer and their association with clinicopathological characteristics [J]. J Zhejiang Univ (Med Sci), 2014, 43(4): 406-412.
[4] CHEN Su-han,et al. Fourier transform infrared spectroscopy in detection of breast cancer [J]. J Zhejiang Univ (Med Sci), 2014, 43(4): 494-500.