首页 | 本学科首页   官方微博 | 高级检索  
检索        

乳腺X线图像纹理分析技术在乳腺结节定性诊断中的应用
引用本文:张海洋,蔺海.乳腺X线图像纹理分析技术在乳腺结节定性诊断中的应用[J].临床医学工程,2021(3):285-286.
作者姓名:张海洋  蔺海
作者单位:南阳市中心医院肌电图室;南阳市中心医院放射治疗科
摘    要:目的分析乳腺X线图像纹理分析技术在乳腺结节定性诊断中的应用价值。方法选择2019年3月至2020年3月我院收治的126例乳腺结节患者,均接受乳腺X线检查,以手术病理组织检查结果为金标准,分析乳腺X线图像纹理分析技术在乳腺结节定性诊断中的应用价值。结果手术病理组织检查结果显示,126例乳腺结节患者中,良性95例,恶性31例;乳腺X线诊断乳腺结节患者的灵敏度、特异度、准确度分别为77.42%、 91.58%、 88.10%;在乳腺X线图像纹理分析定量参数中,恶性组标准差、熵、不均匀度均高于良性组(P <0.05),但两组的平均值比较无统计学差异(P>0.05)。ROC曲线结果显示,标准差、熵、不均匀度及联合诊断预测乳腺结节患者恶性病变的AUC分别为0.724、 0.731、 0.740、 0.811。结论乳腺X线图像纹理分析技术在乳腺结节定性诊断中的应用价值较高,可提升乳腺良恶性病变诊断准确度,为后期治疗提供有效依据。

关 键 词:乳腺结节  乳腺X线  图像纹理分析技术  定性诊断

Application of Breast X-Ray Image Texture Analysis Technique in the Qualitative Diagnosis of Breast Nodules
ZHANG Haiyang,LIN Hai.Application of Breast X-Ray Image Texture Analysis Technique in the Qualitative Diagnosis of Breast Nodules[J].Medical and Health Care Instruments,2021(3):285-286.
Authors:ZHANG Haiyang  LIN Hai
Institution:(Department of Electromyography,2 Department of Radiotherapy,Nanyang Central Hospital,Nanyang 473000,China)
Abstract:Objective To analyze the application value of breast X-ray image texture analysis technique in the qualitative diagnosis of breast nodules.Methods 126 patients with breast nodules admitted to our hospital from March 2019 to March 2020 were selected and given breast X-ray.Taking the results of surgical pathological examination as the gold standard,the application value of breast X-ray image texture analysis technique in the qualitative diagnosis of breast nodules was analyzed.Results The results of surgical pathological examination showed that Among 126 patients with breast nodules,95 cases were benign and 31 cases were malignant.The sensitivity,specificity and accuracy of breast X-ray in the diagnosis of breast nodules were 77.42%,91.58%and 88.10%,respectively.In the quantitative parameters of breast X-ray image texture analysis,the standard deviation,entropy and non-uniformity of malignant group were higher than those of benign group(P<0.05),but no statistical difference was found in the average value between the two groups(P>0.05).ROC curve showed that the standard deviation,entropy,non-uniformity and combined diagnosis of AUC for predicting malignant lesions of breast nodules were 0.724,0.731,0.740 and 0.811,respectively.Conclusions Breast X-ray image texture analysis technique in the qualitative diagnosis of breast nodules has higher application value,which can improve the diagnostic accuracy of benign and malignant breast lesions,and provide effective basis for later treatment.
Keywords:Breast nodule  Breast X ray  Image texture analysis technique  Qualitative diagnosis
本文献已被 维普 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号