论文部分内容阅读
目的探讨超声弹性成像整体应变率比值法与局部应变率比值法诊断乳腺良恶性病灶的最佳临界点,并比较2者的鉴别诊断价值。方法对78例患者共88个乳腺病灶分别行整体应变率比值与局部应变率比值测量,以病理诊断为金标准,构建受试者工作特征(ROC)曲线,分别获得2种方法鉴别乳腺病灶良恶性的最佳临界点,并比较其诊断价值。结果构建ROC曲线后,整体应变率比值的曲线下面积(AUC)为0.90,确定的最佳诊断界点为3.03,其敏感性、特异性、准确性、阳性预测值、阴性预测值分别为92.0%、81.6%、86.3%、86.5%、86.1%;局部应变率比值的曲线下面积(AUC)为0.96,确定的最佳诊断界点为30.89,其敏感性、特异性、准确性、阳性预测值、阴性预测值分别为94.0%、92.1%、93.2%、94.0%、92.1%,局部应变率比值诊断乳腺病灶良恶性的效能高于整体应变率比值(P<0.05)。结论整体应变率比值法和局部应变率比值法对乳腺病灶良恶性的鉴别诊断具有较大价值,后者优于前者。
Objective To investigate the best critical point for the diagnosis of benign and malignant breast lesions by the ratio of the whole strain rate of ultrasound elastography and the local strain rate ratio, and to compare the differential diagnostic value of the two methods. Methods A total of 88 breast lesions were examined in 78 patients with ratio of overall strain rate to local strain rate. The pathological diagnosis was taken as the gold standard, and the receiver operating characteristic (ROC) curve was constructed. Two methods were respectively used to identify the breast lesions The best critical point of malignancy, and compare its diagnostic value. Results After constructing the ROC curve, the area under the curve (AUC) of the overall strain rate ratio was 0.90 and the best diagnostic cutoff point was 3.03. The sensitivity, specificity, accuracy, positive predictive value and negative predictive value were 92.0 %, 81.6%, 86.3%, 86.5% and 86.1%, respectively. The area under the curve of local strain rate (AUC) was 0.96. The optimal diagnostic cutoff point was 30.89. The sensitivity, specificity, accuracy and positive predictive value The values of negative predictive value and negative predictive value were 94.0%, 92.1%, 93.2%, 94.0%, 92.1% respectively. The ratio of local strain rate in diagnosing benign and malignant breast lesions was higher than the overall strain rate (P <0.05). Conclusion The overall strain rate ratio method and the local strain rate ratio method have great value in the differential diagnosis of benign and malignant breast lesions, the latter is better than the former.