This study proposes an optimized ensemble learning approach with explainable AI (XAI) for improved heart disease prediction. The approach integrates the robustness of ensemble learning algorithms with the precision of Bayesian optimization for hyperparameter tuning and the interpretability offered by Shapley Additive Explanations (SHAP). The ensemble classifiers considered include adaptive boosting (AdaBoost), random forest, and extreme gradient boosting (XGBoost). The experimental results on the Cleveland and Framingham datasets demonstrate that the optimized XGBoost model achieved the highest performance, with specificity and sensitivity values of 0.971 and 0.989 on the Cleveland dataset and 0.921 and 0.975 on the Framingham dataset, respectively. The study emphasizes the importance of model interpretability in healthcare, where understanding the rationale behind predictions is as crucial as the accuracy. SHAP values are used to explain the model's decision-making process, providing insights into how each feature influences the prediction. The proposed approach is evaluated on two publicly available datasets containing various clinical and demographic features of patients. The performance of the ensemble classifiers with Bayesian optimization is compared with the standard classifiers without optimization. The results show that the optimized XGBoost model outperforms other models in terms of accuracy, specificity, and sensitivity, demonstrating its effectiveness in heart disease prediction. The SHAP analysis reveals key risk factors and their contributions to the predictions, highlighting the model's robustness and clinical relevance. The study concludes that the integration of Bayesian optimization with ensemble learning and SHAP values provides a promising approach for heart disease prediction in clinical settings.This study proposes an optimized ensemble learning approach with explainable AI (XAI) for improved heart disease prediction. The approach integrates the robustness of ensemble learning algorithms with the precision of Bayesian optimization for hyperparameter tuning and the interpretability offered by Shapley Additive Explanations (SHAP). The ensemble classifiers considered include adaptive boosting (AdaBoost), random forest, and extreme gradient boosting (XGBoost). The experimental results on the Cleveland and Framingham datasets demonstrate that the optimized XGBoost model achieved the highest performance, with specificity and sensitivity values of 0.971 and 0.989 on the Cleveland dataset and 0.921 and 0.975 on the Framingham dataset, respectively. The study emphasizes the importance of model interpretability in healthcare, where understanding the rationale behind predictions is as crucial as the accuracy. SHAP values are used to explain the model's decision-making process, providing insights into how each feature influences the prediction. The proposed approach is evaluated on two publicly available datasets containing various clinical and demographic features of patients. The performance of the ensemble classifiers with Bayesian optimization is compared with the standard classifiers without optimization. The results show that the optimized XGBoost model outperforms other models in terms of accuracy, specificity, and sensitivity, demonstrating its effectiveness in heart disease prediction. The SHAP analysis reveals key risk factors and their contributions to the predictions, highlighting the model's robustness and clinical relevance. The study concludes that the integration of Bayesian optimization with ensemble learning and SHAP values provides a promising approach for heart disease prediction in clinical settings.