Predictive System for Heart Disease Using a Machine Learning Trained Model

Abstract

Heart as one of the essential organ of the human body and with its related disease such as cardiovascular diseases accounts for the death of many in our society over the last decades, and also regarded as one of the most life-threatening diseases in the world.  Hence we seek to predict a system for Heart disease using a supervised Machine Learning (ML) trained model in MATLAB2018 workflow in a real-time environment. To develop the system, 299 heart sounds from patients were obtained and labeled as normal and abnormal heart sound. Features were extracted and labeled as dataset; K Nearest Neighbour (KNN), Support Vector Machine (SVM) and Decision Tree (DT) algorithm were used as the training platform. From the classification analysis developed using the supervised ML trained model in MATLAB2018 in conjunction with system software features for the prediction of the heartbeat for both current and predefined of a heart condition algorithms used in training the dataset for the prediction when principle component analysis was enabled, the result shows that KNN algorithm has the highest and best accuracy of 94.4%, followed by the SVM with 84.4% and DT had 81.1%.  while from the evaluation analysis, KNN on Receive Operation Characteristic Curve (ROC) with 90% variance and training time of 12.88 seconds on positive class of abnormal over false classes of normal heart sound has AUC as 0.94 and on ROC curve with PCA 90% variance and training time of 1.7119 seconds on positive class of normal over negative classes of abnormal heart sound has AUC as 0.89 efficiency.

Hence the analysis from the result shows that out of the three classified algorithms used, KNN predicts and have the highest accuracy and is more efficient with respect to real-time environment.