A machine learning model and application for heart disease prediction using prevalent risk factors in Nigeria
Keywords:
machine learning, supervised learning machine learning algorithm, heart disease prediction, prevalent risk factorsAbstract
Over the past two decades, heart disease remains the principal source of death globally. According to the World Health Organization, about 17.9 million deaths occur every year internationally as a result of heart disease. The vast number of deaths is common amongst low and middle-income countries and Nigeria is at the top on the list of the most affected subSaharan African Nations. With the ravaging impact of heart disease in developing countries, there is need to have a reliable, accurate and efficient approach to make an early diagnosis of the disease to achieve prompt therapy or management. The purpose of this research is to develop a machine learning model and application for heart disease prediction using prevalent heart disease risk factors in Nigeria. Quantitative, qualitative, and experimental research methods were used in implementing this research. The experiment for the training and testing of the models were carried out using Decision Tree (DT), Random Forest (RF), Naïve Bayes (NB) and Voting Ensembles Classifier (VEC) with NB, DT, and RF as base learners. The DT, NB, RF and VEC machine learning algorithms gave an accuracy and error rate values of (95.83%, 0.042), (97.22%, 0.028), (97.92%, 0.021) and (96.53%, 0.035) respectively. This research developed a heart disease prediction model and application using RF and Django python web framework respectively. The
heart disease dataset used in training the model was gotten from Federal Medical Centre Keffi, Nassarawa State, Nigeria. The developed system acts as a decision support tool for cardiologists
