DMHZ: A Decision Support System Based on Machine Computational Design for Heart Disease Diagnosis Using Z-Alizadeh Sani Dataset

Abstract

Cardiovascular heart disease is at the top of the list of lethal diseases which causes major cancer deaths worldwide with mortality rate of approximately 7 million per annum. The development of machine computational design would help in early detection, prognostication and timely diagnosis of disease which will help in increasing the life-span of a patient. In the proposed work, a framework based on machine computations, DMHZ, is proposed for heart disease diagnosis which is validated using Z-Alizadeh Sani heart disease dataset from UCI repository. DMHZ utilizes the feature extraction techniques principal component analysis (PCA) for numeric feature extraction and multiple correspondence analysis (MCA) for categorical feature extraction. The model, DMHZ, is trained using machine learning classifiers, Logistic Regression (LR), Random Forest (RF) and Support Vector Machine (SVM), and validated using holdout validation scheme with hold-out ratio 3:1. Experimentation results show that DMHZ outperforms several state-of-the-art methods in terms of accuracy.