Blockchain-aided comparative study of heart disease detection using machine learning-based approaches with an...

Heart disease, also known as cardiovascular disease (CVD), is a diverse set of conditions that disrupt the normal functioning of the cardiovascular system by narrowing the coronary arteries. These arteries are used for blood circulation and the delivery of essential nutrients and oxygen to various bodily parts. Heart disease accounts for approximately 31% of global deaths, amounting to over 17.9 million lives lost annually. This staggering global death toll due to cardiovascular disease underscores the critical imperative for comprehensive research and innovative solutions. In response to this urgent need, this study employs a multifaceted approach to address the detection of cardiovascular disease, leveraging a combination of advanced techniques. Specifically, it integrates machine learning models for disease prediction, a robust encryption algorithm for data security, and a private blockchain framework for ensuring tamper-proof storage, transparency, and data integrity. Together, these components form a comprehensive system designed to accurately detect cardiovascular conditions and safeguard sensitive medical information. Here, a voting ensemble technique and an array of machine learning models, including decision trees, random forests, Naïve Bayes, k-nearest neighbors (KNN), and neural networks, were used and achieved an accuracy of 89.2%. As a consequence, this study presents an enhanced approach aimed at mitigating the profound impact of cardiovascular disease. Its implementation promises improved individual health outcomes and more effective management of the societal and economic challenges inherent in cardiovascular disease.