Chronic Kidney Disease (CKD) is a significant global health concern characterized by major global health issues defined by progressive kidney damage. Early detection is essential for chronic kidney disease (CKD) as it often progresses silently and can lead to severe complications. This study introduces a novel hybrid optimization algorithm, combining the Bat Algorithm (BA) and Grey Wolf Optimizer (GWO), to address the challenges of CKD prediction. By leveraging BA’s exploration capabilities and GWO’s exploitation potential, the proposed algorithm effectively searches for optimal solutions, enhancing the accuracy and efficiency of CKD diagnosis. The proposed algorithm, termed BAGWO, iteratively refines solutions through a two-stage process, enhancing the model’s ability to converge on accurate and informative results. This algorithm not only enhanced classification accuracy but also reduced the feature set to only six attributes. The Random Forest classifier resulted in a recognition rate of 99.5 %, making it the top-performing classifier.