Fig. 4

Summary of SHAP values for the top 20 features in the training set, internal test set, and external test set. (A), (B), and (C) represent the training set, internal test set, and external test set, respectively. The color in the plot indicates the size of each feature’s value, with the color gradient from blue to red representing an increase in variable value from low to high. The horizontal axis of the points represents the SHAP values for the corresponding features, reflecting the extent of the feature’s impact on the model’s predictions. Positive SHAP values indicate a positive effect, while negative SHAP values indicate a negative effect.