Cardiovascular Disease Prediction using Patient History and Real Time Monitoring
K Ananthajothi, Joshua A. David, A Kavin
Abstract
A comprehensive healthcare solution through a unified web application is offered Centered on cardiovascular disease prediction and broader health prognosis based on patient treatment history and recent health data, it integrates three distinct datasets and deploys top-tier machine learning algorithms tailored to each dataset's unique objectives. For cardiovascular disease prediction, Random Forest, Support Vector Machine, and Gradient Boosting algorithms are employed, meticulously analyzing patient health records and lifestyle factors to accurately identify disease risks. In the domain of disease prediction using health data, Logistic Regression, Long Short-Term Memory networks, and Multilayer Perceptrons provide insights into various health conditions by assessing recent health metrics. The third dataset, which tracks individuals' activity and health metrics over a month, relies on K- Means Clustering, Decision Trees, and Recurrent Neural Networks to monitor activity patterns and deviations, issuing timely health warnings to users. This all-inclusive web application unites predictive power with real-time health monitoring, empowering individuals to proactively manage their well-being, fos tering a culture of preventive care, and facilitating personalized health management in a user-friendly and accessible online platform.