Utilizing Machine Learning Approaches for Cardiovascular Disease Forecasting
Keshika Seeboruth, Lai Zhi Wen, Vazeerudeen Abdul Hameed, Au Yit Wah, Kesava Pillai Rajadorai, Muhammad Ehsan Rana
Abstract
Heart disease, often referred to as cardiovascular disease, encompasses various conditions impacting the heart and has consistently been the top cause of death worldwide for many years. The disease is linked with numerous risk factors, underscoring the importance of methods that are precise, trustworthy, and swift for early detection and timely treatment. Data mining has gained prominence in the healthcare field as a way to handle massive data volumes. Scholars employ a range of data mining and machine learning methods to interpret intricate medical data, aiding healthcare providers in forecasting heart disease incidents. In this empirical study, we examine multiple factors related to heart disease and construct a model using supervised learning algorithms such as Logistic Regression, Random Forest, and Support Vector Machine. We trained these models on the Framingham Heart study dataset, which is openly accessible on Kaggle. This dataset contains 4239 records with 16 characteristics. The main objective of this paper is to predict the probability of individuals developing coronary heart disease within ten years. Our results suggest that the Random Forest algorithm delivers the most accurate performance in this predictive task.