Skip to main content
. 2023 May 24;14:1151728. doi: 10.3389/fimmu.2023.1151728

Figure 2.

Figure 2

Machine learning model performance. Linear regression analysis. (B) XGBoost model: feature importance. (C) Model performance. Receiver-operating characteristic curves for 5 machine learning models. The XGBoost model achieved a larger (better) AUROC compared with the other models: (D) Train ROC curve, (E) Validation ROC curve. (F) Forest plot of the AUC Score of the 5 models. (G) Calibration plots of 5 models. The XGBoost achieved lower (better) Brier scores compared with the other models. (H) Decision curve analysis for machine learning models. (I) SHAP analysis was performed on the XGBoost model to visually represent the importance of each feature. Each feature is represented by a color that corresponds to the variable’s value, with red indicating a larger value and blue indicating a smaller value. This analysis provides insight into the relationship between each feature and its importance in the model. (A).