Fig. 4
From: Novel nomogram and decision curve analysis for predicting head and neck skin cancer risk
Performance verification plots of the simplified model (after excluding reverse causality variables). (A) Receiver Operating Characteristic (ROC) curve of the simplified nomogram in the training cohort, with an Area Under the Curve (AUC) of 0.850 (95% confidence interval [CI]: 0.824–0.876); (B) ROC curve of the simplified nomogram in the validation cohort, with an AUC of 0.865 (95% CI: 0.826–0.905); (C) Calibration curves of the simplified nomogram in the training cohort, including apparent (red solid line), ideal (black dashed line), and bias-corrected (green solid line) curves (illustrating the agreement between predicted and actual risks); (D) Calibration curves of the simplified nomogram in the validation cohort (curve definitions consistent with Panel C); (E) Decision Curve Analysis (DCA) of the simplified nomogram in the training cohort; (F) DCA of the simplified nomogram in the validation cohort (curve definitions consistent with Panel E).
