Table 3.
Sensitivity and specificity of seven predictive models in the training, test and validation sets.
| Algorithmic models | Training set |
Test set |
Validation set |
|||
|---|---|---|---|---|---|---|
| Sensitivity | Specificity | Sensitivity | Specificity | Sensitivity | Specificity | |
| DL | 95.90% | 98.08% | 80.59% | 83.33% | 89.83% | 87.61% |
| GBM | 92.91% | 94.57% | 85.57% | 84.62% | 84.75% | 95.58% |
| XGBoost | 93.66% | 87.54% | 88.06% | 84.62% | 94.92% | 83.19% |
| XRT | 85.82% | 90.10% | 83.58% | 88.46% | 94.92% | 89.61% |
| DFR | 82.46% | 86.90% | 83.58% | 84.62% | 93.22% | 91.15% |
| GLM | 83.96% | 81.79% | 79.10% | 84.61% | 81.36% | 82.30% |
| Stacked Ensemble | 97.76% | 94.89% | 80.60% | 87.18% | 92.33% | 92.92% |