Heart Disease Prediction Using Hybrid Machine Learning: A Brief Review

Abstract

Cardiovascular disease is a widespread and potentially fatal condition that requires proactive preventive measures and efficient screening approaches on a global scale. To tackle this issue, recent studies have investigated novel machine-learning frameworks that propose to diagnose and forecast cardiovascular disease by capitalizing on enormous datasets and predictive patterns linked to this condition. The research contribution is a thorough examination and implementation of ensemble learning and other hybrid machine-learning techniques for heart disease prediction. By employing ensemble learning on datasets including The Cleveland heart disease dataset and The IEEE Dataport heart diseases dataset such as age, chest pain type, blood pressure, blood glucose level, ECG in rest, heart rate, and four types of chestpain. To predict heart disease, our methodology integrates numerous machine learning models. By capitalising on the merits of specific algorithms while addressing their drawbacks, this approach yields a predictive model that is more resilient. The findings of our research exhibit encouraging outcomes in the realm of heart disease prediction, attaining enhanced precision and dependability in contrast to discrete algorithms. Through the utilisation of ensemble learning, we successfully discerned predictive patterns that are linked to heart disease, thereby augmenting the capabilities of diagnostics. In summary, the findings of our study emphasise the considerable potential of ensemble techniques within the realm of machine learning for the advancement of cardiac disease prediction. By providing a more dependable method for rapid diagnosis and prognosis of cardiac disease, this strategy has substantial ramifications for healthcare practices.