Machine Learning Models and Medical Data Analysis: An Experiment

Cardiovascular diseases (CVD) still constitute the leading causes of death and morbidity throughout the world, becoming more concerned with public health problems, especially in low- and middle-income countries (LMICs). If we can detect heart conditions early, then we would be able to decrease the mortality rates caused by them. The accurate diagnosis of heart diseases is challenging particularly because that testing demands considerable expertise, time, and knowledge from healthcare professionals. This case of the women should remind us that we need more AI powered technologies to risk-stratify these high-risk cardiovascular patients, and provide them a means to efficiently predict the future outcome. In the event high risk populations can be identified, early interventions such as changes in diet and statin therapy may be used to advance primary prevention of CVD. The main goals of this study were to evaluate six machine learning models K-Nearest Neighbors (KNN), Support Vector Classifier (SVC), XGBoost, Logistic Regression (LR), Decision Tree, and Random Forest (RF). The leader of the field was KNN, with an accuracy of around 86% followed by the SVC model. However, other models also have produced nearly equal results on the same dataset.