Heart Disease Prediction and Detection Using Machine Learning

Abstract

The critical need for efficient early-stage identification is underscored by the fact that cardiovascular diseases (CVDs) continue to rank as the top cause of mortality worldwide. It is very difficult for doctors to make an accurate diagnosis of early-stage cardiac disease. The good news is that there are now viable options for rapid diagnosis and treatment thanks to developments in current diagnostic technology. The overarching goal of this study is to draw connections between the capabilities of ML and DL algorithms for cardiovascular health data analysis. Improving the precision and consistency of HD prediction models is the main goal. This study adds to the growing body of knowledge on the use of categorization and predictive analytics in healthcare, with an emphasis on cardiovascular diseases. Various approaches are suggested to accomplish this goal, including the use of AI techniques, a hybrid DL methodology (RNN+GRU) for HD prediction, and a Soft Voting Ensemble (SVE) ML methodology. The study makes use of a number of machine learning classifiers, such as RF, LR, NB, KNN, GB, LGB, and AB. Precision rate, Area Under Curve (AUC), F1 Score, sensitivity, and accuracy are the metrics used to assess the proposed system