Transformative Approaches in Early Cardiovascular Risk Evaluation through AI-Enhanced Modeling Techniques

The prevalence of cardiovascular disease (CVD) around the globe is steadily increasing. Thus, there is an increasing need to implement ways to assess and identify Individual Risk for CVD in Proactive ways. To assess this risk, The Behavioral Risk Factor Surveillance System (BRFSS) 2022 data from available Public Health Surveys have been utilized to develop The Artificial Intelligence (AI) Heart Disease Risk Assessment System, in order to better predict the Future Cardiovascular Disease Risk through the identification of Cardiovascular Risk factors early on, in order to improve patient outcomes and delay disease progression. The research program utilizes AI to identify those that are at Risk of developing heart disease, through the use of (but not limited to) supervised machine learning algorithms, (Random Forests, Light Gradient Boosting Machines and Logistic Regression etc.), and UAA (Universal Applicability of Algorithms). The Entire Data Science Pipeline (Data Wrangling, Distribution-Based Imputation, EDA (Exploratory Data Analysis), Advanced Feature Selection & Robustly Developed Machine Learning Models) has, also, been implemented into this study and the proposed system is a good example of how explainable Artificial Intelligence and accessible Digital Platforms can work together to help early Identify Cardiovascular Health Risks and Support Preventive Health Behaviors.[1] Also, to increase prediction accuracy, the framework utilizes Feature Engineering, Distribution-Based Imputation and Thorough Preprocessing of Data. Overall, the results indicated that The Easy Ensemble Classifier (with Light) provides the best and most reliable model for large, real-time screening of cardiovascular risk by providing the most balanced level of Sensitivity and Specificity.