Fig. 5

The reasonable analysis of the established clinical prediction model. (A) Receiver Operating Characteristic (ROC) diagram of reasonable analysis in the training set. (B) ROC diagram of reasonable analysis in the validation set. The results showed that the Area Under the Curve (AUC) of the nomogram model is higher than that of single-variable models in both the training set and the validation set. This indicates that the predictive performance of the six-variable model is superior to that of any single-variable model. (C) Decision curve analysis (DCA) diagram of reasonable analysis in the training set. (D) DCA diagram of reasonable analysis in the validation set. The results of the DCA diagram of reasonable analysis showed that the net benefit rate of the nomogram model is higher than that of a simple model with thresholds ranging from 0.1 to 0.8 in the training set and from 0.3 to 0.8 in the validation set. This indicates that the net benefit of the six-variable model is superior to that of single-variable models in both the training and validation sets. (E) The results of the model score comparison the Significant differences in nomogram scores between the SSI and Non-SSI groups in the training set. (F) The results of the model score comparison the Significant differences in nomogram scores between the SSI and Non-SSI groups in the validation sets. The results of the model score comparison indicating the model’s effectiveness. P values were calculated via two independent samples t-tests. SSI surgical site infection, BMI body mass index, PSD postoperative suction drainagen, GS gelatin sponge, NPTA none-preoperative antibiotic, TT thrombin time.