A Review on Cardiovascular Disease Detection using Machine Learning Algorithms

Abstract

Cardiovascular Disease (CD) is a type of disease that includes Heart or Blood vessels related problems. CD includes Coronary Artery Disease (CAD) such as angina, heart attack, congenital heart disease etc. CD detection and segmentation from cardiac anatomy is needed in the initial stage to prevent disease. The whole heart is segmented by using medical image models like Magnetic Resonance Imaging (MRI) and Computed Tomography (CT) plays a vital role in diagnosis and treatment. Cardiac Magnetic Resonance Imaging (MRI) gives information about diagnosis of cardiovascular by providing parameters like thickness of myocardium, Left Ventricle (LV) volume and ejection fraction of Right Ventricle (RV). Cardiac MRI segmentation has become a leading medical imaging issue. Existing methods used for cardio vascular segmentation give overfitting problems and analysis of short-axis and long-axis CMR images was used but it will not work with imbalance dataset. Deep Convolutional Neural Networks (CNN) have made success in cardiovascular image segmentation but larger training datasets and imbalance datasets will not give optimum results so further improvements are required. This paper presents a wide range of surveys on how machine learning algorithms are beneficial in predicting Cardiovascular Disease and provides the comparison among those algorithms in terms of efficiency and accuracy.

Keywords-Cardiovascular Disease Detection, Computed Tomography, Deep Convolutional Neural Network, Magnetic Resonance Imaging, Overfitting