Comparative Analysis of Machine Learning and Deep Learning Models for Cardiovascular Disease Prediction: An Artificial Neural Network Approach

Abstract

Cardiovascular diseases remain a leading cause of global morbidity and mortality, and early detection through non-invasive means is critical for improving patient outcomes. In this study, we present a comparative analysis of machine learning and deep learning models for cardiovascular disease prediction using a dataset of 1,025 anonymized patient records from public cardiovascular health databases, each containing 14 demographic, physiological, and clinical attributes. The data preprocessing pipeline included mean imputation for numerical features, mode imputation for categorical variables, and Min-Max normalization, with the final dataset split into 80% training and 20% testing sets. We implemented five classifiers—Logistic Regression, Decision Tree, Random Forest, Support Vector Machine, and an Artificial Neural Network (ANN)—and evaluated their performance using accuracy, precision, recall, and F1-score. The ANN achieved the highest predictive performance with an accuracy of 93.8%, precision of 93.1%, recall of 94.2%, and F1-score of 93.6%, followed by the Random Forest model at 91.3% accuracy. Furthermore, the ANN attained an Area Under the Receiver Operating Characteristic Curve of 0.96, indicating outstanding discriminatory capability and minimal false-positive and false-negative predictions. This research demonstrates that deep learning models significantly outperform traditional machine learning algorithms for this classification task. Therefore, the proposed ANN-based framework offers a reliable and efficient decision-support tool for early cardiovascular disease screening in clinical settings, contributing to timely interventions and improved healthcare delivery. The findings underscore the potential of integrating advanced artificial intelligence into routine diagnostic workflows