This paper presents an interpretable and explainable AI approach for heart stroke prediction, addressing the gap between complex ML models and their clinical interpretability. The study uses the Stroke Prediction Dataset, which contains 11 attributes, to develop a model that incorporates techniques such as resampling, data leakage prevention, and feature selection. The proposed model aims to enhance both accuracy and interpretability, making it more accessible to healthcare practitioners. Key contributions include the use of permutation importance and LIME (Local Interpretable Model-Agnostic Explanations) to provide global and local explanations of the model's predictions. The Artificial Neural Network (ANN) model achieved an accuracy rate of 95%, demonstrating the effectiveness of the proposed approach. The research also discusses the importance of feature selection and data preprocessing techniques, such as handling missing values and class imbalance, to improve model performance. The study concludes with a comprehensive evaluation of the model's performance using various metrics and highlights the potential of the proposed method to improve patient outcomes and healthcare practices.This paper presents an interpretable and explainable AI approach for heart stroke prediction, addressing the gap between complex ML models and their clinical interpretability. The study uses the Stroke Prediction Dataset, which contains 11 attributes, to develop a model that incorporates techniques such as resampling, data leakage prevention, and feature selection. The proposed model aims to enhance both accuracy and interpretability, making it more accessible to healthcare practitioners. Key contributions include the use of permutation importance and LIME (Local Interpretable Model-Agnostic Explanations) to provide global and local explanations of the model's predictions. The Artificial Neural Network (ANN) model achieved an accuracy rate of 95%, demonstrating the effectiveness of the proposed approach. The research also discusses the importance of feature selection and data preprocessing techniques, such as handling missing values and class imbalance, to improve model performance. The study concludes with a comprehensive evaluation of the model's performance using various metrics and highlights the potential of the proposed method to improve patient outcomes and healthcare practices.