Table 3.
Metrics of predictive models. Se—sensitivity, Sp—specificity, Ka—kappa (normalized proportion of correct answers), Pr—precision, AUC—area under the error curve.
| Models | Test Dataset/Training Dataset | ||||
|---|---|---|---|---|---|
| Se | Sp | Ka | Pr | AUC | |
| Based on clinicopathological parameters | 0.67/0.75 | 0.61/1.00 | 0.60/0.93 | 0.27/1.00 | 0.631/0.875 |
|
ALDH3A2 + CHKA + ODF2 + QSOX2 + hsa-miR-503-5p |
1.00/1.00 | 0.93/1.00 | 0.94/1.00 | 0.75/1.00 | 0.963/1.000 |
|
ALDH3A2 + CHKA + ODF2 + QSOX2 + hsa-miR-503-5p + ISUP |
1.00/1.00 | 0.96/1.00 | 0.97/1.00 | 0.86/1.00 | 0.982/1.000 |