声触诊组织量化成像技术对乳腺BI-RADS4类病变的诊断价值

Diagnostic Value of Virtual Touch Imaging Quantification for BI-RADS 4 Breast Lesions

[目的]评估声触诊组织量化成像（vitual touch imaging quantification,VTIQ）对BIRADS 4类乳腺病变的诊断价值。[方法 ]分析150例BI-RADS 4类乳腺病变的VTIQ值,以病理结果为标准,绘制ROC曲线,确定良恶性诊断的VTIQ最佳阈值,校正BI-RADS分类,分析VTIQ校正前后诊断率有无统计学差异。[结果]150个病变中良性76个,恶性74个。BI-RADS分类诊断乳腺病变良恶性ROC曲线下面积为0.807,敏感度、特异性、准确率、阳性预测值和阴性预测值分别为81.58%（62/76）、74.32%（55/74）、78.0%（138/150）、76.54%（62/81）和79.71%（55/69）。乳腺恶性病变的SWV值明显高于良性病变（P〈0.001）,诊断良恶性的最佳阈值SWV为4.09m/s,校正后ROC的曲线下面积为0.884,敏感度、特异性、准确率、阳性预测值和阴性预测值分别为90.79%（69/76）、90.79%（69/74）、92%（138/150）、93.24%（69/74）和90.79%（69/76）。[结论 ]VTIQ有助于诊断BI-RADS 4类乳腺病变,优化穿刺适应证的选择。

[Objective]To evaluate the application of virtual touch imaging quantification（VTIQ）in the diagnosis of BI-RADS 4 breast lesions.[Methods]One hundred and fifty patients with pathologically diagnosed breast lesions were enrolled in the study.All patients underwent ultrasonographic examination and the results were analyzed with VTIQ software.With pathologic results as the reference standard,the receiver-operating characteristic（ROC）curve with BI-RADS classification and with VTIQ correction were developed and the diagnostic value for breast lesions before and after correction was analyzed.[Results]Among 150 cases of breast lesions,there were74 benign lesions and 76 malignant lesions.The VTIQ values of malignant lesions were significantly higher than those of benign lesions（P0.001）.The area under the ROC curve（AUC）of BIRADS classification was 0.807 and the AUC after correction was 0.884.Taking VTIQ of 4.09 m/s as cut-off value,the sensitivity,specificity,accuracy,positive predictive value and negative predictive value of BI-RADS for diagnosis of malignant breast lesions were 81.58%（62/76）,74.32%（55/74）,78.0%（138/150）,76.54%（62/81）and 79.71%（55/69）,respectively.The sensitivity,specificity,accuracy,positive predictive value,negative predictive value of VTIQ correction were90.79%（69/76）,90.79%（69/74）,92%（138/150）,93.24%（69/74）and 90.79%（69/76）,respectively.[Conclusion]Virtual touch imaging quantification technique can improve the diagnosis of BIRADS 4 classes breast lesions.