PURPOSE:To explore the application value of SMI scoring assignment method combined with 2017 American College of Radiology (ACR) Thyroid Imaging, Reporting and Data System (TI-RADS) in differentiating benign and malignant thyroid nodules. METHODS:According to the 2017 ACR TI-RADS classification, the enrolled nodules were divided into 3 points group, 4 points group, 5 points group, 6 points group and≥7 points group. The nodules were assigned scores according to the echocity of the nodules and the microvessels detected by SMI and their distribution patterns based on ACR TI-RADS. Accompany with the scores increased or decreased after assignment, the thyroid nodules were re-grouped. RESULTS:The AUC after the scores assignment is better than before (Z = 3.881, P < 0.001). The specificity, positive predictive value and accuracy after score assigned are better than those of before (Z = 8.323, P < 0.001; Z = 8.619, P < 0.001; Z = 5.345, P < 0.001), there is no statistical difference in sensitivity before and after score assigned (Z = –0.513, P = 0.60), and the negative predictive value before assigned score is better than that of after (Z = –3.826, P < 0.001). CONCLUSION:The diagnostic efficacy after scoring was better than that of before.

中文翻译：

---

超微血管成像（SMI）评分分配法在常规超声鉴别甲状腺良恶性结节中的价值

目的：探讨SMI评分分配法结合2017美国放射学会（ACR）甲状腺影像、报告和数据系统（TI-RADS）在鉴别甲状腺结节良恶性中的应用价值。方法：根据2017年ACR TI-RADS分级，纳入的结节分为3分组、4分组、5分组、6分组和≥7分组。根据 SMI 检测到的结节和微血管的回声及其分布模式，根据 ACR TI-RADS 对结节进行评分。随着分配后评分的增加或减少，甲状腺结节被重新分组。结果：评分分配后的 AUC 优于分配前（Z = 3.881，P < 0.001）。特异性，评分后阳性预测值和准确率均优于前（Z = 8.323，P < 0.001；Z = 8.619，P < 0.001；Z = 5.345，P < 0.001），前后敏感性无统计学差异分配得分（Z = –0.513，P = 0.60），分配得分前的阴性预测值优于分配得分后的阴性预测值（Z = –3.826，P < 0.001）。结论：评分后诊断疗效优于评分前。