An efficient system for the prediction of coronary artery disease using dense neural network with hyper parameter tuning

Abstract

Diagnosis of heart disease is considered as one of the challenging problems in medical science in the current decade. Coronary artery disease is a type of heart disease in which the arteries of the heart gets affected. Hence many researchers propose a number of intelligent solutions to improve the predictability towards the identification of Coronary artery disease. If the disease can be identified at an early stage, then precautions can be taken for its recovery. In the proposed system, an efficient deep learning technique is used for improving accuracy towards the identification of the disease. The proposed system is built using a Dense Neural Network which is a type of deep learning network. Here the experimentation is done using Cleveland Heart disease data set present in the UCI repository. The system has three stages. In the first stage data cleaning and feature selection is performed. In the second stage model training is done using hyper parameter tuning. In the last stage, the trained model is used for prediction of coronary artery disease using test data set. The proposed model results in the classification accuracy of 96.03% during training and an accuracy of 94.91% during testing, which is best among all the discussed methods.