Fig. 6

A. SHapley Additive exPlanations (SHAP) values for clinical features in predicting the need for surgery (surgery required). Each dot represents a SHAP value for a single patient’s feature, with its position indicating the degree of influence on the model’s decision and its color representing the feature’s value (red = high, blue = low). B. SHAP values for clinical features in predicting the need for surgery (surgery not required), following the same format as Fig. 6A, illustrating how these features influenced cases where surgery was not required. C. Mean absolute SHAP values across both classes, offering a global perspective on feature importance in the model’s decision-making