Abstract
<title>Abstract</title> <p>Cardiovascular disease (CVD) continues to be one of the major global causes of death; therefore, the early prediction of CVD is crucial for proper healthcare delivery. In this paper, an AI-based multi-parametric network system based on the clinical and lifestyle information of almost 300 patients was designed. The dataset included several important risk predictors, such as age, blood pressure, smoking and alcohol drinking, stress level, physical activity, and prior existing conditions of the patient. After performing the data cleaning in order to address the problem of missing values, exploratory data analysis and the creation of correlation heat maps were used to determine connections between different clinical parameters. Logistic Regression, Decision Tree and Random Forest machine learning models were built and assessed. The highest accuracy results were achieved by Decision Tree and Random Forest classifiers with 94% accuracy in both cases. Cytoscape-based networks were created in order to analyze interactions between the major CVD risk predictors.</p>