A Comparative Analysis of Machine Learning Models for Heart Disease Prediction Using Clinical Parameters

Abstract

Cardiovascular disease remains one of the leading causes of mortality worldwide, with early detection being crucial for improving patient outcomes. This study aims to develop and validate a machine learning-based prediction model for heart disease using electronic health records from a major hospital system in New Jersey. We will analyze retrospective data from adult patients who received care between 2021 and 2024, incorporating demographic information, vital signs, laboratory values, medication history, and imaging results. The study will utilize supervised machine learning algorithms, including random forest, gradient boosting, and deep learning approaches, to predict the presence of significant coronary artery disease. The dataset will be split into training (70%) and validation (30%) sets, with feature selection performed to identify the most significant predictors of cardiovascular disease. Performance metrics including accuracy, sensitivity, specificity, and area under the receiver operating characteristic curve will be calculated to evaluate model effectiveness. The findings from this study will contribute to the development of more accurate risk prediction tools for cardiovascular disease in clinical settings and potentially aid in early intervention strategies. Furthermore, this research will provide insights into the most influential predictors of heart disease specific to our patient population in New Jersey.