扩散加权成像图像纹理分析鉴别乳腺浸润性导管癌与纤维腺瘤的初步研究  

Texture analysis for differentiating breast invasive ductal carcinoma and breast fibroadenoma

作　　者：王清霖 毛宁[2] 史英红[2] 马恒 董建军[2] 王滨 郝翠娟 刘梅婕 车凯莉 张学喜 谢海柱[2] WANG Qinglin;MAO Ning;SHI Yinghong;MA Heng;DONG Jianjun;WANG Bin;HAO Cuijuan;LIU Meijie;CHE Kaili;ZHANG Xuexi;XIE Haizhu(Department of Medical Imageology,School of Clinical Medicine,Binzhou Medical University,Yantai 264000,China;Department of Radiology,the Affiliated Yantai Yuhuangding Hospital of Qingdao University,Yantai 264000,China;GE Healthcare China,Shanghai 250000,China)

机构地区：[1]滨州医学院临床医学院医学影像学系,山东烟台264000 [2]青岛大学附属烟台毓璜顶医院影像科,山东烟台264000 [3]GE Healthcare China,上海250000

出　　处：《实用放射学杂志》2018年第9期1368-1371,1422,共5页Journal of Practical Radiology

基　　金：中国医药教育协会专项课题（2016SKT-M034）；山东省自然科学基金项目（ZR2017PH043）；山东省医药卫生科技发展计划项目（2016WSA06057）。

摘　　要：目的 探讨扩散加权成像（DWI）图像纹理分析在鉴别乳腺浸润性导管癌与乳腺纤维腺瘤的可行性。方法 选取经手术病理证实的，在术前行MRI检查的38例患者的DWI图像，共40个病灶，其中乳腺浸润性导管癌20例，乳腺纤维腺瘤20例，分别对其DWI图像进行纹理特征分析，并结合受试者工作特征曲线（ROC曲线）评估2组间纹理参数的鉴别诊断价值，采用Fisher判别模型计算2类乳腺肿瘤鉴别的准确率。结果 统计得到方差、标准差、强度及熵值等67种纹理特征参数中19个纹理特征参数在2组数据中均有统计学意义（P〈0.05），利用ROC曲线发现平均偏差及相对偏差在浸润性导管癌及乳腺纤维腺瘤鉴别中具有较高诊断价值，同时采用Fisher判别模型对2类乳腺肿瘤鉴别的准确率为92.5％，行交叉验证鉴别准确率为85.0％。结论 浸润性导管癌及纤维腺瘤在DWI图像的纹理特征显著不同，DWI的纹理分析在两者鉴别方面具有可行性，且诊断效能高于常规DWI，可以作为鉴别两者的辅助方法。Objective To determine the feasibility of using texture analysis to differentiate breast invasive ductal carcinoma and fibroadenoma on diffusion weighted imaging （DWI）.Methods This study included 38 patients who underwent breast MRI prior to operation and had pathological findings, with a total of 40 breast tumors including 20 cases of breast invasive ductal carcinoma and 20 cases of breast fibroadenoma. The textural features of DWI images were analyzed. The receiver operating characteristic curve （ROC curve） was calculated to evaluate the diagnostic efficiency of texture parameters.Fisher discriminant model was used to determine the efficiency of texture parameters for distinguishing the two types of breast tumors.Results A total of 19 texture feature parameters, such as variance, standard deviation, intensity, and entropy, out of 67 texture parameters were statistically significant in the two sets of data （P〈0.05）. By comparing ROC curves, we found that the mean and relative deviations exhibited high diagnostic values in differentiating between invasive ductal carcinoma and fibroadenoma. The accuracy of Fisher discriminant analysis for the two types of breast tumors was 92.5%,and the accuracy rate of cross validation was 85.0%.Conclusion The texture characteristics on DWI images of invasive ductal carcinoma and fibroadenoma are significantly different. Texture analysis based on DWI images is feasible in identifying them, and the diagnostic efficiency is higher than that in conventional DWI.It can be used as an auxiliary method to identify them.

关 键 词：乳腺 浸润性导管癌 纤维腺瘤 纹理分析 扩散加权成像 

分 类 号：R737.9[医药卫生—肿瘤] R445.2[医药卫生—临床医学]