The performance metrics of model 6 using TFSP in the derivation and validation cohorts
| Data set | Classifier | Accuracy | AUC | SPE | SEN | Precision | F1 | Log |
| Derivation | RF | 0.77 | 0.85 (0.78–0.93) | 0.72 | 0.83 | 0.78 | 0.80 | 0.48 |
| Validation | RF | 0.75 | 0.80 (0.73–0.87) | 0.71 | 0.78 | 0.81 | 0.75 | 0.55 |
AUC, area under the curve; Log, log loss; RF, random forest; SEN, sensitivity; SPE, specificity; TFSP, two-stage feature selection pipeline.