Heat diseases prediction using machine learning

Computer Engineering

Heart disease is a critical global health issue and remains one of the leading causes of mortality, particularly among middle-aged and elderly populations. Traditional diagnostic methods such as angiography and stress testing, while effective, are often invasive, expensive, and not always accessible, especially in under-resourced regions. To address these limitations, this study explores the use of machine learning (ML) techniques to develop a predictive model that can assess the likelihood of heart disease based on clinical and lifestyle-related patient data. This research utilizes a publicly available dataset containing health-related records of over 400,000 individuals from the United States. The study involves the implementation and evaluation of six widely-used machine learning algorithms: XGBoost, Bagging Classifier, Random Forest, Decision Tree, K-Nearest Neighbors (KNN), and Naïve Bayes. Each model is trained and tested using standard performance evaluation metrics, including accuracy, precision, recall, F1-score, and ROC-AUC, to determine their effectiveness in predicting heart disease. Among all the models evaluated, the XGBoost classifier demonstrated the highest predictive performance, achieving an accuracy of 91.30%. The superior results are attributed to XGBoost’s ability to handle complex feature interactions and its robustness against overfitting. This study emphasizes the potential of ML-driven approaches in building scalable, accurate, and cost-effective diagnostic tools that can assist healthcare providers in early detection and personalized risk assessment. By integrating such models into healthcare systems, this research aims to support timely clinical decision-making and ultimately contribute to reducing the global burden of heart disease.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.