Mohamed Nadjib Boufenara, Mahmoud Boufaida And Mohamed Lamine Berkane, Constantine 2 Abdelhamid Mehri University, Lire Laboratory, Constantine, Algeria
Heart disease is one of the most fatal diseases in the world. The difficulty of the diagnosis and the need to identify the risk at an early stage required the construction of an effective prediction system. Machine learning is powerful for building prediction systems. However, some classification algorithms achieve good prediction accuracy, while others predict with low accuracy. In this article, we propose a deep learning model that allows decision support for the diagnosis of heart disease in patients. The proposed model is compared to four machine learning models which are: Random Forest, Logistic Regression, Naive Bayes and SVM. In order to assess the performance of the models, we used the dataset of patients suffering from coronary heart disease, which comes from a cardiovascular study in Massachusetts residents. The deep learning model was able to achieve the best accuracy compared to other models.
Coronary Heart Disease, Deep Learning, Machine Learning, Classification.
Copyright © BIEN 2020