Integrating Clinical Assessment Indicators into Cardiovascular Risk Event Simulation Using Machine Learning and Agent Based Modeling

Cardiovascular disease (CVD) remains the leading global cause of death, with approximately 17.9 million mortalities annually. Studies have shown that adopting healthy behaviors, i.e., a balanced diet, regular physical activity, and weight management, can reduce CVD risk. However, evaluating their long-term impact requires extensive data collection and analysis, which are both time-consuming and challenging. This study developed a novel mathematical framework integrating an agent-based model (ABM) to simulate CVD risk progression and established clinical guidelines into synthetic training data for machine learning (ML) classification. The ML model was trained entirely on synthetic data generated from World Health Organization/International Society of Hypertension cardiac risk indications, and validated using outcomes from a NetLogo simulation. The workflow does not use real patient data; instead, the expected simulation results serve as a reference to assess the ML model and synthetic data. The ABM, designed in NetLogo, exchanges agent characteristics with a trained ML model to classify individuals into appropriate CVD risk levels based on lifestyle and clinical parameters. The simulation indicated measurable risk progression (5–12%) by year 20 in individuals with both smoking and diabetes. A combined effect of high dietary intake and low physical activity showed over 20% risk increase, demonstrating the model’s capacity to capture dynamic risk interactions. The relationship between CVD risk and systolic blood pressure was also effectively reproduced. Additional scenarios confirmed the alignment of model outcomes with real-world trends, showing model self-consistency, identifying critical thresholds and population-level risk shifts through detailed tabular analysis. Beyond confirming known associations, the findings support the internal consistency of the model, highlighting its potential as a simulation based tool for studying cardiovascular risk patterns and supporting risk monitoring within controlled settings.