Fig. 8

Development and evaluation of a nomogram model based on three independent prognostic factors. a ROC curve used to evaluate the 1-, 3-, and 5-year predictive efficiency of the five CpG sites model in the training set. b ROC curve used to evaluate the 1-, 3-, and 5-year predictive efficiency of the five CpG sites model in the validation set. c ROC curve used to evaluate the 1-, 3-, and 5-year predictive efficiency of the nomogram model. d Nomogram model constructed by using three independent prognostic factors including risk score model, age, and tumour grade. e Calibration curve of the risk score model. f Calibration curve of the nomogram model. g, h Decision curve analysis between the risk score and nomogram model. g The net benefit of the risk score model was lower than that of the nomogram model. h The net reduction in interventions per 100 patients of the nomogram model was higher than that of the risk score model