Performance Evaluation of Machine Learning Algorithms for Coronary Artery Disease Features
Jalal, Md. Shah
MetadataShow full item record
Cardiovascular disease is the leading cause of mortality in the world. Bangladesh probably has the highest rates of cardiovascular disease among all South Asian countries and yet is the least studied. A proper prediction mechanism system can significantly reduce this death toll. In this work, we propose an intelligent system that can make an effectively prediction of a possible cardiac attack using only twelve (12) features. We also apply seven (06) well known supervised machine learning algorithms on two different datasets (e.g. NICVD patient’s data and UCI dataset) to analyze the prediction accuracy. The overall process can be categorized into four phases. Phase 1: we have provided a comprehensive literature review where we summarize various related machine learning algorithms. Phase 2: we have collected heart disease patients’ data through survey questionnaires from NICVD, Bangladesh to create a dataset. Phase 3: we have reduced feature vector dimensionality with heart disease prediction model. Finally, feed the data to appropriate machine learning algorithms to determine if the predictive model is accurate. It is observed that using NICVD patient’s data for 12 features the classification accuracy of Artificial Neural Network (ANN) is 92.80% and it performs better than others classification algorithms such as Decision Tree (82.50%), Naïve Bayes (85%), Support Vector Machine (SVM) (75%), Logistic Regression (77.50%), Random Forest (75%). Whereas using UCI dataset with 12 attributes the classification accuracy of ANN is 91.7% and it also performs better than the other classification algorithms, such as Decision Tree (76.90%), Naïve Bayes (86.50%), SVM (76.33%), Random Forest (67.33%), and Logistic Regression (81.52%). In Bangladesh most of the people don't go for regular medical checkup. The main reasons behind this ignorance are lack of money, as the whole medical process is too costly so most people can't afford this, and lack of awareness. Among all diseases, heart attack is most common and hazardous is not only Bangladesh but also universal. Heart attack occurs when stream of blood cannot go to the brain for any blockage. The angiographic heart disease status is costly and also it is not available in rural area. This intelligent system can be predict heart disease status by measuring some simple information that will be helpful for everyone as it will not take much time and will be budget friendly. It can be assist to medical practitioners or noncardiologists, especially in rural area.
- M.Sc Thesis/Project