应用Logistic回归模型筛选剪切波弹性成像评估颈部淋巴结的定量参数

发布时间：2018-05-17 04:40

本文选题：颈部 + 淋巴结　；参考：《中国医学影像技术》2016年04期

【摘要】：目的采用Logistic回归模型探讨剪切波弹性成像(SWE)评估颈部淋巴结良恶性的价值和筛选SWE的定量参数。方法对59例患者共95个疑似恶性颈部淋巴结在颈部淋巴结清扫前进行常规超声检查及剪切波弹性成像检查,分别比较颈部良性与恶性淋巴结的弹性值比值(E-ratio)、病灶的平均弹性值(E-mean)、病灶的最大弹性值(E-max)和标准差(SD);以病理结果为金标准,建立二分类逻辑回归模型,绘制模型预测概率的ROC曲线并计算曲线下面积,确定诊断界值,计算敏感度、特异度、准确度、阳性预测值、阴性预测值和约登指数。结果Logistic回归模型为logitic(p)=-3.653+1.760 X_1+0.235 X_2-0.207 X_3+0.168 X_4,X_1为E-ratio,X_2为E-mean,X_3为E-max,X_4为SD。模型预测概率ROC曲线的曲线下面积为0.865,以55.66%为模型预测概率的诊断界值时,准确率最高为84.21%,对应的敏感度为80.00%,特异度88.89%,阳性预测值为88.89%,阴性预测值为80.00%,约登指数为68.89%。结论运用SWE的4个定量参数建立的逻辑回归模型对颈部良恶性淋巴结的鉴别具有中等诊断价值,4个定量参数均可为良恶性淋巴结的鉴别诊断提供依据,而诊断价值最高的定量参数是E-ratio。
[Abstract]:Objective to evaluate the value of shear wave elastography (SWE) in evaluating benign and malignant cervical lymph nodes by using Logistic regression model and to screen quantitative parameters of SWE. Methods Ninety-five suspected cervical lymph nodes of 59 patients were examined by conventional ultrasound and shear wave elastography before neck lymph node dissection. The ratio of elastic value of benign and malignant cervical lymph nodes was compared, the mean elastic value of the lesion was E-mean, the maximum elastic value of the lesion was E-maxand, and the standard deviation of SDN was compared, and a two-classification logical regression model was established based on the pathological results as gold standard. The ROC curve of the prediction probability of the model was drawn and the area under the curve was calculated. The diagnostic threshold value, sensitivity, specificity, accuracy, positive predictive value, negative predictive value and Jordan index were determined. Results the Logistic regression model is logitic(p)=-3.653 1.760 XS 1 0.235 X2-0.207 XSerge 3 0.168 X 4 XS 1, E-ratio X 2, E-meann X3, E-maxX 4, SD4. The area under the curve of model prediction probability ROC curve is 0.865. When 55.66% is the diagnostic threshold value of model prediction probability, the highest accuracy is 84.21, the corresponding sensitivity is 80.00,the specificity is 88.89, the positive predictive value is 88.89, the negative predictive value is 80.00,and the Yorden index is 68.89. Conclusion the logical regression model based on the four quantitative parameters of SWE is of moderate value in the differential diagnosis of benign and malignant lymph nodes in the neck, and the four quantitative parameters can provide the basis for the differential diagnosis of benign and malignant lymph nodes. The most valuable quantitative parameter was E-ratio.
【作者单位】：中山大学孙逸仙纪念医院超声科;
【基金】：国家自然科学基金面上项目(30872996)
【分类号】：R445.1;R730.44

【相似文献】