Explainable AI-Based Clinical Decision Support for Knee Osteoarthritis Surgery Prediction

Knee osteoarthritis (OA), one of the most common degenerative joint disorders affecting the elderly population, is a source of pain, disability, and decreased quality of life. A portion of individuals need surgical intervention, such as total knee arthroplasty, as the illness worsens. For better clinical results and individualized treatment planning, it is essential to determine which patients are most likely to require surgery. In order to forecast whether patients with osteoarthritis in their knees would require surgery, this study suggests a clinical decision support framework based on explainable artificial intelligence (XAI) and organized clinical data. Several machine learning models were created and assessed using accuracy, F1-score, and area under the receiver operating characteristic curve. The Random Forest classifier outperformed the models in prediction, demonstrating its ability to capture non-linear correlations within clinical variables. SHapley Additive exPlanations (SHAP) were incorporated into the framework to address the lack of transparency connected to machine learning models. SHAP analysis offered the significance of global features and patient-specific justifications. The suggested approach improves clinical trust by combining interpretability with prediction accuracy. This study demonstrates the potential of explainable AI-driven decision support systems to facilitate surgical decision-making in clinical practice.