CareSync: An AI-Powered Big Data Health Monitoring System for Early Disease Prediction and Personalized Recommendations

The rise in chronic health conditions, such as cardiovascular diseases, obesity, diabetes, and stress-related disorders, highlights the need for ongoing, smart health monitoring solutions. Traditional healthcare methods often struggle due to fixed appointment schedules and slow diagnostics, making early intervention tough. Current models mainly use basic machine learning techniques and often do not effectively handle complex, large-scale health data or adjust in real time. In response, this study introduces CareSync, a health monitoring framework that uses big data technologies and machine learning to predict health risks and provide personalized health recommendations. The system gathers biometric data from wearable devices and Google Fit, processes it using PySpark, and employs models like Random Forest and XGBoost for classification. Auto ML model has been used for improving the predictive performance which automates the model tuning and selection. In evaluations, this optimized model showed impressive results, achieving a PR-AUC of 96.7% and a ROC-AUC of 99.99%. Standalone models, such as XGBoost and Random Forest, recorded accuracies of 86% and 84%, respectively. CareSync’s main contribution is its combination of scalable data processing with AI-driven risk assessment and personalized recommendations, making it a strong tool for early disease detection and better patient self-management. By integrating artificial intelligence, big data frameworks, and model optimization, CareSync offers a significant advantage over traditional systems by providing timely, user-specific insights to support preventive healthcare.