Logistic回归模型评价声触诊组织成像和定量技术在甲状腺结节鉴别诊断中的价值

目的建立以甲状腺结节超声诊断特征为变量的回归模型,评价甲状腺影像报告和数据系统(TI-RADS)中的5项超声特征及声触诊组织成像和定量技术(VTIQ)反映的结节硬度在甲状腺结节良恶性鉴别诊断中的价值。方法对95例甲状腺结节患者的120个结节行常规超声及VTIQ检查,以病理诊断为金标准建立Logistic回归模型。评价Logistic回归模型的预报能力,计算各变量的似然比,评价弹性与常规超声各项超声特征在甲状腺结节良恶性鉴别诊断中的价值,然后将原分类系统中预测能力最差的一项由弹性特征替代,绘制ROC曲线,比较校正前后ROC曲线下面积。结果运用Logistic回归分析,筛选出对良恶性鉴别诊断中有统计学意义的甲状腺结节超声特征,包括微钙化、边缘不清及结节质地硬。Logistic回归模型对甲状腺结节良恶性预报的准确率为80.83%(97/120),敏感度为75.93%(41/54),特异度为84.85%(56/66)。弹性替代实质结节重新评估甲状腺结节分类的ROC曲线下面积分别为0.700和0.797,P0.05。结论VTIQ反映的弹性特征较常规超声中的实质结节特征更有助于甲状腺结节良恶性的鉴别诊断。

Evaluation of virtual touch tissue imaging quantification technology in differential diagnosis of thyroid nodule

Objective To assess the value of thyroid imaging reporting and data system (TI-RADS) and virtual touch tissue imaging quantification ( VTIQ ) in the differential diagnosis of thyroid nodule by applying the binary logistic regression model . Methods Conventional ultrasound and VTIQ of 120 thyroid nodules confirmed by surgical pathology in 95 patients were retrospec-tively analyzed. A binary logistic regression model was obtained based on the stiffness and the five features of TI-RADS. The odds ratios of ultrasonographic malignant features were calculated and the value of stiffness and TI-RADS features for differentiating be-nign and malignant nodules was compared. The less useful TI-RADS feature was replaced by the stiffness and receiver operating characteristic (ROC) curves were drawn to compare the correction. Results Stiffness and TI-RADS features of microcalcification and irregular margin were included into the logistic model. The accuracy, sensitivity and specificity were 80.83%, 75.93% and 84.85%, respectively. The areas under the ROC curves increased significantly (P<0.05) when TI-RADS features of microcalcifica-tion (0.700) and irregular margin (0.797) were replaced by stiffness. Conclusion VTIQ is more useful than TI-RADS features in the differential diagnosis of thyroid nodule.