Advancing Cardiovascular Disease Prediction: An Interpretive Evaluation of Machine Learning and Deep Learning Models

Abstract

The heart is an important organ required for human beings. It is also the main organ of the cardiovascular system which pumps the blood to the body. The Cardiovascular disease (CVD) is the dominant cause of human death worldwide. Therefore, there is an urgent need to develop precise and practical predictive tools for understanding and diagnosing this disease in advance. Owing to advancements in Machine Learning (ML) and Deep Learning (DL), the accuracy of CVD prediction has significantly increased. Therefore, it offers a groundbreaking potential for early disease identification and provides individual treatment for patients. In this study, the prominent models used in ML such as Logistic Regression (LR), Decision Trees (DT), Random Forest (RF), Support Vector Machine (SVM), and K-nearest neighbors (KNN), and important DL models in DL such as artificial neural networks (ANN), convolutional neural networks (CNN), and recurrent neural networks (RNN), were implemented, compared, and critically analysed. By comparing and analysing different existing models, it is proposed that, by using multimodal data and hybrid models, the accuracy can be increased to the next highest benchmark. The paper concludes that the RF and DT models performed extraordinarily well with an accuracy of 98.5% for the dataset used in the present study.