Cardiovascular Disease Detection using Artificial Immune System and other Machine Learning Models

Abstract

Researchers and medical institutions face problems in detecting and diagnosing cardiovascular diseases in the early stages. Therefore, having a tool for detecting cardiovascular diseases in early stages will be helpful for the medical institutions to combat the disease. In this paper, we have presented a solution for detection of cardiovascular diseases by using clonal selection algorithm. Clonal selection is an Artificial Immune system (AIS) based algorithm which is often used for pattern recognition problems. Here, we propose modified clonal selection algorithm (CLONALG) which effectively detects cardiovascular diseases. With the proposed algorithm, we have achieved an average accuracy of 78%. Further, we compared the accuracy of CLONALG algorithm with different models of Machine Learning, viz. Random Forest Classifier (RFC), Decision Tree Classifier (DTR), Support Vector Machines (SVM), Logistic Regression (LR) and Artificial Neural Networks (MLP-ANN) for cardiovascular disease detection.