超声联合纹理分析对乳腺结节良恶性的诊断价值

目的以乳腺结节的灰阶超声及剪切波弹性超声影像为基础,分析超声图像的纹理特征,探索常规超声联合纹理分析对乳腺结节良恶性的诊断价值。 方法前瞻性收集2018年8月至2018年12月于首都医科大学附属北京友谊医院常规超声检查发现乳腺结节并获得病理诊断的患者113例,共113个结节。所有患者均行常规超声及剪切波弹性成像检查,并对113个乳腺结节依据乳腺影像报告和数据系统（BI-RADS）进行分类;对超声图像进行纹理分析,获得纹理特征参数并建立诊断模型。以病理结果为"金标准",分析纹理特征诊断模型、常规超声联合纹理特征诊断模型对乳腺结节良恶性的诊断价值。 结果以乳腺结节穿刺病理结果为"金标准",纹理分析诊断乳腺结节良恶性的敏感度、特异度、阳性预测值、阴性预测值及准确性分别为0.64、0.91、0.75、0.86、0.83,ROC曲线下面积为0.77;常规超声与纹理分析联合方法诊断乳腺结节良恶性的敏感度、特异度、阳性预测值、阴性预测值及准确性分别为0.88、0.85、0.71、0.94、0.86,ROC曲线下面积为0.86。联合诊断的ROC曲线下面积高于纹理分析,差异有统计学意义（Z=2.133,P=0.03）。采用常规超声联合纹理分析方法,90.0%（72/80）的BI-RADS 4a类乳腺结节可以降级为BI-RADS 3类,病理结果显示,降级为BI-RADS 3类的乳腺结节中94.4%（68/72）为良性结节。 结论常规超声联合纹理分析对乳腺结节良恶性有较好的诊断效能,可减少不必要的有创性检查,具有良好的应用前景。

Diagnostic value of ultrasound combined with texture analysis in differentiating benign and malignant breast nodules

ObjectiveTo explore the diagnostic value of routine ultrasonography combined with texture analysis in breast nodules based on their gray-scale ultrasound and shear-wave elastography images. MethodsFrom August 2018 to December 2018, 113 patients with 113 breast nodules were prospectively included, and the breast nodules were found by routine ultrasonography and diagnosed by pathology. All patients underwent conventional ultrasound and shear wave elastography, and 113 breast nodules were classified according to Breast Imaging Reporting and Data System (BI-RADS). Texture feature parameters were obtained through texture analysis of ultrasound images and a diagnostic model was established. Using pathological results as the "golden standard", the diagnostic value of the texture feature diagnostic model and routine ultrasonography combined with the texture feature diagnostic model for benign and malignant breast nodules was analyzed. ResultsUsing the pathological results of puncture specimens of breast nodules as the "golden standard", the sensitivity, specificity, positive predictive value, negative predictive value, and accuracy of texture analysis in the diagnosis of benign and malignant breast nodules were 0.64, 0.91, 0.75, 0.86, and 0.83, respectively, and the area under the ROC curve (AUC) was 0.77. The sensitivity, specificity, positive predictive value, negative predictive value, and accuracy of routine ultrasonography combined with texture analysis in the diagnosis of benign and malignant breast nodules were 0.88, 0.85, 0.71, 0.94, and 0.86, respectively, and the AUC was 0.86. The AUC of the combined diagnosis was significantly higher than that of texture analysis (Z=2.133, P=0.03). Based on the combined diagnosis, 90.0% (72/80) of BI-RADS 4a breast nodules could be degraded to BI-RADS 3. Pathological results showed that 94.4% (68/72) of the breast nodules degraded to BI-RADS 3 were benign. ConclusionConventional ultrasound combined with texture analysis has better diagnostic efficacy for benign and malignant breast nodules. The combined diagnosis can reduce needless puncture and has good application prospects.