An Exhaustive Comparative Analytical Study of 15 Machine Learning Models For Automated Cardiovascular Disease Classification

Cardiovascular diseases are increasingly becoming a silent killer with over 17.9 million deaths annually due to ischaemic heart disease and strokes. Additionally, that number is set to rise at a rapid pace as obesity rates and unhealthy eating habits increase worldwide. Cardiovascular diseases are quite well hidden among individuals and mostly remain undetected until they strike, with a high fatality rate. An active testing protocol must be established to ensure that all such problems are detected and mitigated at an earlier stage in order to improve the quality of life and life expectancy of individuals suffering from cardiovascular diseases.This generates a large patient dataset, and to accurately prepare models to handle the automated detection, they have to be trained on an existing data set which was chosen to a combination of 4 existing data sets from the USA, Hungary, and Switzerland, where each has 13 features and 1 target class which needs to be accurately analyzed and predicted to ensure that large scale checkups can occur without burdening the healthcare system. In the present study, the authors have worked to perform a series of exploratory data analysis, and base line predictions via 15 varying Machine Learning models to understand the type of models that perform better and can be utilized for further optimization.