Our research addresses the pressing issue of congestive heart failure (CHF), a critical cardiovascular condition characterized by the heart’s diminished ability to pump blood effectively, resulting in fluid accumulation. Current diagnostic methods often face challenges in signal processing and manual Electrocardiogram (ECG) analysis, leading to reduced accuracy and diagnostic complexities. To tackle these challenges, we introduce an innovative framework that integrates global QRS average (gQRS) detection of RR peaks and intervals from ECG data. We then apply advanced machine learning models such as K-Nearest Neighbours (KNN), Support Vector Machine (SVM), Random Forest (RF), and XGBoost, specifically tailored for CHF diagnosis. What distinguishes our approach is the strategic use of ensemble learning, combining the predictive strengths of XGBoost and RF algorithms. This fusion optimizes diagnostic outcomes and demonstrates significant improvements in CHF detection accuracy, marking a notable advancement in clinical diagnostics. Our research underscores the potential of ensemble learning methodologies in enhancing diagnostic accuracy and clinical decision-making for CHF. By leveraging cutting-edge technologies and methods, we aim to revolutionize cardiovascular health monitoring and contribute to more effective patient care strategies. This innovative approach not only achieves a high accuracy rate of 99.56% but also significantly reduces processing time, making our research highly impactful and promising for practical healthcare applications.