Advancing Heart Disease Detection by using Ensemble Meta-Features Integration

heart disease continues to be a prominent contributor to both illness and death on a worldwide scale, presenting a substantial obstacle to public health. Heart disease can manifest in various forms, each impacting the cardiovascular system differently. This paper presents a novel and efficient approach for heart disease detection using a ML-based ensemble model with integrated meta-features. The proposed model combines five distinct techniques: Weighted Average Ensemble, Adaptive Boosting, Random Forest with Feature Bagging, Negative Correlation Learning, and Ensemble of Small Models. Each technique contributes to the overall predictive power of the model, leveraging diverse strategies to enhance accuracy and robustness. To compare the proposed model, traditional ML algorithms are also applied and compared the results with proposed model. The experiments conducted demonstrate the efficiency of the proposed ensemble learning approach, showcasing its superiority in accurately predicting heart disease. The integration of meta-features enhances the model’s interpretability and adaptability, making it a promising solution for real-world applications in healthcare.