Skip to main content
. 2026 Jan 16;12:1682178. doi: 10.3389/fsurg.2025.1682178

Figure 2.

Graphical representations of model performance and calibration. Panel A: ROC curve with AUC of 0.771 for a model. Panel B: Calibration plot showing predicted versus actual probabilities, with C statistic and various metrics. Panel C: ROC curve with AUC of 0.704. Panel D: Similar calibration plot with different parameters. Panel E and F: Decision curve analyses comparing model performance at different thresholds against \"All\" and \"None\" strategies, highlighting net benefit relative to risk thresholds.

Performance of the frailty prediction model for PAD. (A) ROC curve of the frailty prediction model in the training set. (B) Calibration curve of the frailty prediction model in the training set. (C) ROC curve of the frailty prediction model in the validation set. (D) Calibration curve of the frailty prediction model in the validation set. (E) DCA of the frailty prediction model in the training set. (F) DCA of the frailty prediction model in the validation set.