From prediction to clinical decision-making: Explainable machine learning in acute suppurative cholecystitis

Figure 1 Integration of explainable machine learning into clinical decision-making in acute suppurative cholecystitis. Clinical and imaging data are combined through radiomic feature extraction and integrated into a machine learning model. Explainability methods, such as SHapley Additive exPlanations, provide both global and patient-specific interpretation of model predictions. These outputs support risk stratification and guide clinical decision-making, including surgical prioritization and perioperative management. Multidisciplinary team involvement ensures that predictions are translated into coordinated clinical action, while outcome feedback enables continuous model refinement. SHAP: SHapley Additive exPlanations; CT: Computed tomography; AI: Artificial intelligence.