A KNOWLEDGE BASED FRAMEWORK TO PREDICT CORONARY HEART DISEASE

Abstract

Coronary heart disease (CHD) is a disease common to both men and women and also use of genetics have been rarely used to predict it. Coronary heart disease alludes to a narrowing of the coronary veins, the veins that supply oxygen and blood to the heart. It is otherwise called coronay artery disease. It is a note worthy reason for ailment and death. The existing Clinical Decision Support Systems (CDSSs) have not been accurate enough in their prediction and diagnosis of coronary heart disease. By using genetic information such as Single Nucleotide Polymorphism, Genome Build, Chromosome, Map and the partition coefficient (LogP) gotten from the Duke 2007 dataset and the C4.5 decision tree pattern classification algorithm which was selected amongst other competing classification algorithms including K-Nearest Neighbor, Bayes Classifier and Support Vector Machine after a thorough evaluation on the Waikato Environment for Knowledge Analysis (WEKA version 3.6.7), this study developed a framework for accurate prediction of coronary heart disease. A prediction accuracy of 61.0734% was obtained from training the C4.5 algorithm on the Duke 2007 dataset which gives higher prediction accuracy than the existing CHD Modeling and Execution framework. An improved framework that enhances the classification/prediction of coronary heart disease which helps to guide patients with Cronary heart disease as to how to best manage their health condition and live a normal life.