Table 6. Comparison of the diagnostic ability between ADCb0-800-2000 (mm2/sec) and US in differentiating PTMCs from benign thyroid nodules.
| Imaging diagnosis * | Pathological results | AUC | Sensitivity (%) | Specificity (%) | PPV (%) | NPV (%) | FPR (%) | FNR (%) | Youden's index | ||
|---|---|---|---|---|---|---|---|---|---|---|---|
| PTMCs (n = 33) |
Benign nodules (n = 62) |
||||||||||
| ADCb0-800-2000, mm2/sec | PTMCs | 31 | 9 | 0.940 | 93.94 | 85.48 | 77.50 | 96.36 | 14.52 | 6.06 | 0.79 |
| Benign nodules | 2 | 53 | |||||||||
| US | PTMCs | 33 | 20 | 0.839 | 100 | 67.74 | 62.26 | 100 | 32.26 | 0 | 0.68 |
| Benign nodules | 0 | 42 | |||||||||
* According the ROC analysis, ADCb0-800-2000 ≤ 1.16 mm2/sec was provided as a cutoff value for diagnosing malignant thyroid nodules (PTMCs) and ADCb0-800-2000 > 1.16 mm2/sec for diagnosing benign nodules. Based on the previously published US criteria of TI-RADS, nodules with ≥ 4a were diagnosed as suspicious malignancies (PTMCs) and nodules with 2 or 3 as benign nodules. Abbreviations: ADC, apparent diffusion coefficient; US, ultrasound; PTMCs, papillary thyroid microcarcinomas; PPV, positive predict value; NPV, negative predict value; FPR, False-positive rate; FNR, False-negative rate.