Predicting Heartbeats: Intersection of Machine Learning and Cardiac Health using Ensemble Learning

Cardiac arrest, a sudden and life-threatening medical emergency, claims approximately 17 million lives worldwide annually. Many of those affected, including young individuals and fitness enthusiasts, never anticipate such an event. This lack of awareness, compounded by the inconvenience and cost of traditional medical tests, contributes to the alarming statistics.This research presents three distinct classifiers—Hist Gradient Boosting, Support Vector Classifier (SVC) and Stochastic Gradient Descent Classifier(SGD)—that were implemented, optimized, and evaluated. Furthermore, it explores an ensemble model combining these classifiers using soft voting to maximize predictive accuracy. By analyzing a diverse dataset encompassing patient demographics, medical histories, and clinical parameters, the model aims to provide timely identification for effective recovery and empower healthcare providers.The standout feature of our approach is the development of a user-friendly web portal. This portal allows individuals to assess their cardiac risk from the comfort of their homes in just a few minutes, significantly reducing the time and cost associated with traditional methods. This accessibility encourages proactive health management, potentially reaching a broader audience.