Application for Healthcare Management and Heart Disease Prediction Powered by Machine Learning

Cardiovascular disease (CVD) stands as the foremost global cause of mortality, accounting for over 19.8 million deaths in the year 2022 alone. The challenges of limited access to healthcare and inadequate management of health data frequently result in delayed diagnoses and an increased risk of mortality. In this study, we introduce an Android-based healthcare application that incorporates machine learning (ML) for the early prediction of 10-year coronary heart disease (CHD) risk and the management of patient health. The application boasts a comprehensive data pipeline (including imputation and scaling) and utilizes classification models such as Logistic Regression, Random Forest, and XGBoost, which are integrated through an ensemble stacking method with hyperparameter tuning to improve accuracy. User information (including age, gender, lifestyle, and clinical parameters) is securely managed using Firebase Authentication/Cloud and local SQLite storage, while a medication reminder feature enhances treatment adherence. Evaluations conducted on a publicly available heart disease dataset reveal impressive predictive performance (approximately 85–86% accuracy) with a high AUC-ROC. These findings underscore the potential of AI-driven digital health platforms to facilitate proactive cardiovascular care.