ABSTRACT
People’s health has been found to be more at risk from heart disease, which is one of the major causes of death in the world. According to estimates, 17.9 million deaths worldwide in 2019 were attributed to cardiovascular diseases (CVDs), or 32% of all fatalities. Timely diagnosis and accurate prediction will aid in reducing the mortality rate because of CVDs. Researches are using artificial intelligence (AI) and machine learning (ML) models to effectively diagnose CVDs using structured and ordered datasets (e.g., Cleveland, Framingham and Kaggle). Linear Regression, Naïve Bayes, Support Vector Machine, Decision Tree, and Ensemble Machine Learning Techniques are common techniques that produce comparable results. The aim of using ML in this field is to predict CVDs early, as much as possible, using historical medical data. This study performs a comprehensive literature review by considering works in this field that were published between 2019 and 2022. Different datasets considered by different researchers to predict using ML techniques are outlined in detail. We compared the results of all ML techniques used in these papers.
