COMPARATIVE EVALUATION OF AI MODELS FOR PREDICTING STROKE RISK USING GENETIC AND LIFESTYLE FACTORS

Abstract

Stroke remains a leading cause of disability and mortality worldwide, with genetic and lifestyle factors playing pivotal roles in its onset and progression. This study provides a comprehensive evaluation of various artificial intelligence (AI) models in predicting stroke risk, with a focus on integrating genetic predispositions and lifestyle variables to enhance predictive accuracy. By leveraging data from a large cohort, we examine machine learning algorithms, including decision trees, random forests, neural networks, and ensemble methods, to determine their efficacy in stroke prediction. Each model’s performance is assessed based on accuracy, precision, recall, and F1 score. Additionally, we explore the interpretability of each model, emphasizing the need for transparent AI solutions in healthcare. The findings demonstrate that certain algorithms, particularly ensemble approaches, yield higher predictive accuracy while balancing computational efficiency and interpretability. This research underscores the importance of integrating genetic and lifestyle data in AI-driven health applications, offering significant insights for early intervention and personalized healthcare strategies aimed at stroke prevention.