A Supervised Deep Learning Model for the Detection of Cardiovascular Disease

Abstract

In our world today, cardiovascular disease (CVD) stands as the foremost cause of death worldwide, claiming the lives of nearly 20 million individuals annually. As CVD continues to burden the healthcare industry, there is a critical need for early detection and prevention. The rise of Artificial Intelligence in the medical field offers a range of capable solutions. In order to address the problem, this paper presents the development of a simple, supervised deep learning model for detecting cardiovascular disease in patients. The research focused on creating a model enriched with multiple layers, activation functions, optimizers, and loss functions. The chosen approach leveraged the power of AI to analyze labeled patient data and map input features to corresponding class labels, enabling accurate detection of CVD. The dataset used contained 70,000 patient records with 12 different clinical attributes. In addition, it provides an overview of the most common types of cardiovascular disease, such as coronary artery disease, aortic valve disease, stroke and peripheral artery disease. The accuracy of the obtained results from the deep learning model was up to 73%. The utilization of AI systems can present a novel approach to addressing daily challenges within the rapidly-evolving world of medicine. Health personnel can take advantage of rapidly changing artificial intelligence and user friendly deep learning models to detect similar future medical concerns.