Table 7. Performance estimates obtained from the double cross-validation method.
Outer fold # | Sensitivity | Specificity | Accuracy | p-value |
---|---|---|---|---|
1 | 1.00 | 1.00 | 1.00 | 0.0000001 |
2 | 0.64 | 0.93 | 0.80 | 0.0111700 |
3 | 1.00 | 0.86 | 0.94 | 0.0013510 |
4 | 1.00 | 0.75 | 0.89 | 0.0002533 |
5 | 0.82 | 1.00 | 0.92 | 0.0001199 |
6 | 0.91 | 0.82 | 0.86 | 0.0004277 |
7 | 0.62 | 0.69 | 0.65 | 0.0843200 |
8 | 1.00 | 0.90 | 0.92 | 0.1618000 |
9 | 1.00 | 0.89 | 0.95 | 0.0001114 |
10 | 0.94 | 0.80 | 0.88 | 0.0025660 |
Average | 0.89 | 0.86 | 0.88 | |
Std. Dev | 0.15 | 0.10 | 0.10 |
The performance estimates (sensitivity, specificity, accuracy) for every outer iteration and the corresponding average values are shown. A p-value from McNemar's Chi-square test was computed and p < 0.05 was considered statistically significant.