Fig. 4: Global model explanation by the SHapley Additive exPlanations (SHAP) method for each feature variable within the final model in testing set (n = 59). | Communications Medicine

Fig. 4: Global model explanation by the SHapley Additive exPlanations (SHAP) method for each feature variable within the final model in testing set (n = 59).

From: An interpretable machine learning model for predicting prognosis of medulloblastoma integrating genetic and clinical features

Fig. 4: Global model explanation by the SHapley Additive exPlanations (SHAP) method for each feature variable within the final model in testing set (n = 59).

a Visualization of SHAP summary bar plot depicting the contribution ranking of the XGBoost algorithm’s features in CMR scenario. The bars represent the importance of the variables and their overall contribution to the model prediction. b SHAP summary dot plot. Each patient gets one dot per feature in the model, with the dot color (dark blue = high, light blue = low) showing the actual feature value. Dots stack vertically to show density.

Back to article page