Prioritizing dataset dependability, model performance, and interoperability is a compelling demand for improving stroke risk prediction from medical surveys using AI in healthcare. These collective efforts are required to enhance the field of stroke risk assessment and demonstrate the transformational potential of AI in healthcare. This novel study leverages the CDC’s recently published 2022 BRFSS dataset to explore AI-based stroke risk prediction. Numerous substantial and notable contributions have been established from this study. To start with, the dataset’s dependability is improved through a unique RF-based imputation technique that overcomes the challenges of missing data. In order to identify the most promising models, six different AI models are meticulously evaluated including DT, RF, GNB, RusBoost, AdaBoost, and CNN. The study combines top-performing models such as GNB, RF, and RusBoost using fusion approaches such as soft voting, hard voting, and stacking to demonstrate the combined prediction performance. The stacking model demonstrated superior performance, achieving an F1 score of 88%. The work also employs Explainable AI (XAI) approaches to highlight the subtle contributions of important dataset features, improving model interpretability. The comprehensive approach to stroke risk prediction employed in this study enhanced dataset reliability, model performance, and interpretability, demonstrating AI’s fundamental impact in healthcare.