Decision Curve Analysis (DCA) for the nine predictive models

Decision Curve Analysis evaluating the clinical net benefit of the nine models across different threshold probabilities. ​​Training cohort​​: A: 1 year.B:3 years.C:5 years. ​​Validation cohort​​: D:1 year. E:3 years. F: 5 years. The 'All' and 'None' lines represent the strategies of treating all patients or no patients, respectively. A model with a higher curve indicates greater clinical utility within a specific threshold range. Abbreviations：RSF: Random Survival Forest；GBM: Gradient Boosting Machine; CoxBoost: Cox regression with Gradient Boosting; SurvivalSVM:Survival Support Vector Machine; XGBoost: Extreme Gradient Boosting; SuperPC: Super Principal Component Regression; PLSR-Cox: Partial Least Squares Cox Regression.