The timely and precise detection of multiple sclerosis disease (MS) is crucial for its treatment. This study explores the application of explainable artificial intelligence (XAI) techniques, mainly LIME (Local Interpretable Model-agnostic Explanations) and SHAP (Shapley Additive Explanations) to enhance the analysis of machine learning (ML) models in predicting MS. We developed a predictive model using a comprehensive dataset comprising clinical, demographic, and imaging features. Our analysis revealed that key factors such as age, gender, and specific clinical indicators significantly influence model predictions. SHAP values provided insights into the contribution of individual features, highlighting the importance of age as a critical risk factor for multiple sclerosis. The dependence plots demonstrated a clear correlation between age and model output, reinforcing the need for age-adjusted diagnostic criteria. LIME further augmented our findings by offering localized explanations for individual predictions, thus enhancing the model’s transparency and facilitating clinician trust in AI-assisted decision-making. Moreover, feature reduction, hyperparameter tuning, and stability validation techniques have improved model performance. The results underscore the potential of XAI methods in clinical settings, paving the way for personalized diagnostics and improved patient outcomes in multiple sclerosis. Comparative analysis of proposed models with traditional ML models shows that the XAI models outperformed in accuracy, recall, precision, and F1-score.