剪切波弹性成像技术鉴别良恶性乳腺结节的Logistic回归分析

Logistic regression analysis of shear wave elastography in differential diagnosis of breast lesions

目的:应用Logistic回归模型评价剪切波弹性成像(shear wave elastography,SWE)技术在乳腺结节良恶性鉴别诊断中的价值。方法:对164例患者行常规超声及剪切波弹性成像检查,以病理诊断为金标准建立Logistic回归模型。绘制ROC曲线,评价Logistic回归模型的预报能力,评价剪切波弹性成像技术在乳腺结节良恶性鉴别诊断中的价值。结果:进入回归模型中对乳腺结节良恶性鉴别诊断中有统计学意义的特征变量包括钙化、CDFI、纵横比、弹性模量E max值及E ratio值。Logistic回归模型对乳腺结节良恶性预报的正确率为91.80%,敏感性为93.10%,特异性为89.90%。结论:剪切波弹性成像技术较常规超声检查更有助于乳腺结节的良恶性鉴别诊断。

Objective:To evaluate the value of shear wave elastography in differentiation diagnosis of breast lesions by applying the binary Logistic regression model.Methods:164 cases of breast lesions were examed by conventional ultrasonic and shear wave elastography,which were confirmed by surgical pathology.A Logistic model was obtained on the basis of ultrasonographic features.A receiver operator characters(ROC)curve was constructed to assess the performance of the Logistic model.The differential diagnosis value of shear wave elastography was evaluated.Results:Five ultrasonographic features including calcification,CDFI,aspect ratio,Emax and Eratio were finally entered into the Logistic model.The percentages of correct prediction,the sensitivity and specificity were 91.80%,93.10%and 89.90%,respectively.Conclusion:Shear wave elastography plays more important role in the differential diagnosis of breast lesions.