MACHINE LEARNING–BASED EARLY DETECTION OF CARDIOVASCULAR DISEASE RISK USING EHR AND LIFESTYLE DATA IN PAKISTANI POPULATION COHORTSS

Abstract

Cardiovascular disease (CVD) remains a leading cause of morbidity and mortality worldwide and poses a significant public health challenge in Pakistan. Early identification of individuals at elevated cardiovascular risk is essential for reducing disease burden and improving healthcare outcomes. This study investigates the effectiveness of machine learning techniques for the early detection of cardiovascular disease risk using integrated Electronic Health Records (EHRs) and lifestyle-related data from Pakistani population cohorts. The study employs a predictive healthcare framework incorporating clinical indicators, including blood pressure, cholesterol levels, blood glucose, body mass index, and medical history, alongside lifestyle factors such as smoking behavior, physical activity, dietary patterns, sleep duration, and stress levels. Multiple machine learning algorithms, including Logistic Regression, Random Forest, Support Vector Machine, Artificial Neural Network, and XGBoost, were evaluated to identify the most effective predictive model. The findings indicate that machine learning models significantly improve cardiovascular risk prediction accuracy compared with conventional approaches, with XGBoost demonstrating superior performance. The integration of EHR and lifestyle data substantially enhanced predictive capability and enabled more comprehensive risk assessment. The study contributes to predictive healthcare and cardiovascular informatics literature by providing evidence from a developing-country context and offers practical implications for clinicians, healthcare administrators, and policymakers seeking to strengthen preventive healthcare and data-driven clinical decision-making in Pakistan.