Auto-fill Patient Presets:
Stacked Boosting & Meta-Learner Multi-Disease System
Fine-tuned Stacked Ensemble (XGBoost, LightGBM, CatBoost, GradientBoosting, AdaBoost → Random Forest Meta-Learner)
Discrepancy between training and test sets is minimal, confirming robust generalization.
Shows True Negatives, False Positives, False Negatives, and True Positives on unseen patient test samples.
| Model | Train Acc | Test Acc | F1 | Gap (Overfit) |
|---|---|---|---|---|
| Loading boosting model metrics... | ||||
Adjust dataset training capacity to observe the effect on overfitting and accuracy.
Compute BMI from height and weight