1. ROC Curve
X-axis: 1-Specificity, Y-axis: Sensitivity. AUC closer to 1 is better.
2. Confusion Matrix
Four-fold table and diagnostic metrics (PPV, NPV, LR+, LR-) based on optimal cutoff.
3. Threshold Explorer
Sensitivity, Specificity, and Youden Index vs threshold. Use slider to explore any cutoff.
4. Cutoff Methods
Youden = Sens + Spec - 1. Closest to Top-Left = minimize distance to (0,1). Fixed Sens/Spec = optimize the other metric.
5. Combined Score
When 2+ markers selected, check 'Build Combined Score' to fit logistic regression and use predicted probability as a new marker.
6. Bootstrap
Check to use 1000 bootstrap resamples for AUC CI and DeLong test, more robust for small samples.
7. DeLong Test
Compare AUC differences between markers.
8. Forest Plot
Visual comparison of AUC with 95% CI across all markers.
9. NRI / IDI
Net Reclassification Improvement and Integrated Discrimination Improvement. Compare each marker to the first (reference) marker.
10. Data Export
CSV: BOM header for Windows Excel. Word: A4 landscape with all tables.
11. LR+ / LR-
LR+ = Sensitivity / (1-Specificity). LR- = (1-Sensitivity) / Specificity.
12. PPV / NPV
PPV = TP/(TP+FP). NPV = TN/(TN+FN).
13. Youden Index
Youden = Sensitivity + Specificity - 1.
14. Data Requirements
Outcome: binary. Predictors: continuous numeric. Supports CSV/Excel/TXT.
15. Disclaimer
This tool is for research purposes only. Results should not be used for clinical decision-making without professional validation.