Cardiovascular Disease Detection Using Multiple Machine Learning Algorithms and their Performance Analysis

Heart problems have proven to be lethal all around the world. Cardiovascular diseases like cardiac rhythm disorders, heart failure, congenital heart diseases, etc. are the leading cause of death. In this disease, the heart fails to provide enough blood to other body regions to allow it to perform its regular functions. Cardiovascular disease is detected by traditional invasive procedures such as CT and angiography but they have their limitation to combat such problems and limitations, therefore early and precise diagnosis of this disease is needed for avoiding further damage to patients and protecting their lives in advance. The modern world required intelligent and modern solutions thus in this regard computational strategies built on intelligent machine learning systems have been discovered to be more accurate and effective in the identification of heart disease. This study aimed to develop a system that integrates multiple machine learning algorithms, including K-nearest Neighbor, Naïve Byes, Linear Regression, Decision Tree, and Random Forest, which are used to detect cardiovascular disease. Five machine learning algorithm models were developed and their performances were observed based on several other performance indicators like accuracy, Precision, F1-score, Macro Average, and Weighted average among two target classes i.e. Presence and absence of cardiovascular disease. Classification reports generated against each model were utilized to assess the efficacy and strength of the constructed model.