Table 2.
Performance of ModelU, ModelUR, and ModelURC in different cohorts.
| Cohort | Model | AUC (95% CI) | SEN (%) | SPE (%) | ACC (%) | PPV (%) | NPV (%) |
|---|---|---|---|---|---|---|---|
| Train | ModelU | 0.832 (0.802–0.862)a, b | 68.4a, b | 79.7 | 73.9a, b | 78.1 | 70.4 |
| ModelUR | 0.931 (0.911–0.952)c | 90.2c | 84.8 | 87.6c | 86.3 | 89.1 | |
| ModelURC | 0.955 (0.938–0.971) | 97.7 | 83.1 | 90.6 | 86.0 | 97.2 | |
| Internal validation | ModelU | 0.744 (0.673–0.814)a, b | 66.2a, b | 77.8 | 71.8a, b | 76.1 | 68.3 |
| ModelUR | 0.913 (0.868–0.958)c | 89.6 | 81.9 | 85.9 | 84.1 | 88.1 | |
| ModelURC | 0.934 (0.894–0.974) | 96.1 | 79.2 | 87.9 | 83.1 | 95.0 | |
| External validation | ModelU | 0.637 (0.574–0.693)a, b | 59.3a, b | 62.1a, b | 60.6b | 64.3 | 57.0 |
| ModelUR | 0.842 (0.793–0.885)c | 73.7c | 87.6c | 80.1c | 87.2 | 74.3 | |
| ModelURC | 0.899 (0.861–0.931) | 92.8 | 77.9 | 85.9 | 82.9 | 90.4 |
AUC: area under curve; CI: confidence interval; ACC: accuracy; SEN: sensitivity; SPE: specificity; PPV: positive predictive value; NPV: negative predictive value.
a
Compared the performance of ModelU with the performance of ModelUR, P < 0.05.
b
Compared the performance of ModelU with the performance of ModelURC, P < 0.05.
c
Compared the performance of ModelUR with the performance of ModelURC, P < 0.05.