Analysis of cardiovascular disease using traditional machine learning

The heart is a critical organ in the human body, and any deficiency in its functioning can lead to life-threatening consequences.Cardiovascular diseases are the leading cause of fatality worldwide.Early detection and diagnosis can prevent the risk of life and reduce treatment costs.This paper attempts to investigate a well-defined and utilitarian heart disease research program model based on traditional machine learning models, utilizing datasets from Kaggle and UCI Repository, and determining variables including hypertension, cholesterol level, smoking, sedentary lifestyle, etc.This study uses supervised learning algorithms, including Decision Tree, Random Forest, Logistic Regression, and Support Vector Machine (SVM).The algorithms are evaluated based on their respective performance.Overall, the study showcases the potential of Machine Learning as a decision-making service in health care with timely diagnosis and minimized cost.This study also tests for feature importance, in order to efficacy which clinical and lifestyle aspects would matter throughout the risk of heart disease.If health service organizations can identify which features provide the best information when modelling risk, then they would have the potential to facilitate a priority for patient screening while developing the best individualized treatments.