论文部分内容阅读
目的探讨应用剪切波(SWE)划分TI-RADS 4级的2个亚级,能否改善传统TI-RADS诊断标准的一致性及准确性。方法将240例甲状腺单发实性结节根据TI-RADS诊断标准分为3、4、5级。选取151例TI-RADS 4级者行SWE检查,获取结节杨氏模量平均值,通过与最佳界点比较,将上述4级结节划分为2个亚级,分析其诊断效能及观察者内、观察者间的一致性。另外,两名医师以传统TI-RADS诊断标准划分2个亚级,分析其观察者内、观察者间的一致性及诊断效能。结果 SWE标准观察者内重复性及两名医师间的一致性均较好(k>0.8);而传统TI-RADS诊断标准除高年资医师重复性较好外(k>0.8),其一致性均较差(k<0.8)。SWE标准ROC曲线下面积(AUC)与使用常规TI-RADS诊断标准的高年资医师差异无统计学意义(P=0.734,0.465),而优于同样使用传统TI-RADS标准的低年资医师(P=0.020,0.034)。结论 SWE可作为4级结节2个亚级的判定标准,在不降低准确性的同时可提高其诊断一致性。
Objective To investigate the feasibility and feasibility of using shear wave (SWE) to classify two sub-levels of TI-RADS grade 4 into two groups, which can improve the consistency and accuracy of traditional TI-RADS diagnostic criteria. Methods A total of 240 thyroid solitary solid nodules were divided into 3, 4 and 5 grades according to the diagnostic criteria of TI-RADS. 151 cases of TI-RADS grade 4 were selected for SWE examination to obtain the average Young’s modulus of nodules, compared with the best cut point, the above four nodules were divided into two sub-class, analysis of its diagnostic efficacy and observation Within, the consistency between the observer. In addition, the two physicians were divided into two sub-groups based on the traditional TI-RADS diagnostic criteria, and their intra-observer and interobserver agreement and diagnostic efficacy were analyzed. Results The repeatability of SWE standard observer and the consistency between the two physicians were good (k> 0.8). The traditional TI-RADS diagnostic criteria were the same except that the senior physicians had good repeatability (k> 0.8) Sex is poor (k <0.8). The area under the ROC curve (AUC) for the SWE criteria was not statistically different from those of senior care physicians who used the conventional TI-RADS diagnostic criteria (P = 0.734, 0.465), but was superior to those of the junior physicians who also used the traditional TI-RADS standard (P = 0.020,0.034). Conclusion SWE can be used as the criterion for the determination of grade 2 nodules in grade 4, which can improve the diagnostic consistency without reducing the accuracy.