HeartGuard-AI A Fair, Explainable, and Multimodal Deep Learning Framework for Heart Disease Risk Prediction

The primary objective of this project is to develop a fair, explainable, and multimodal intelligent system for early heart disease prediction. The proposed framework, HeartGuard-AI, integrates three novel deep learning models — FairHeart-Former (Fairness-Aware Transformer), TrustHeart-XAI (Uncertainty-Aware Explainable AI), and MedFusion-HeartNet (Multimodal Clinical Fusion Network) — along with a robust ensemble model. The system classifies individuals into Low, Moderate, and High cardiac risk levels while ensuring fairness across demographic groups, providing model explanations through SHAP, and quantifying prediction uncertainty. A secure and user-friendly web application built using the Flask framework with MySQL authentication has been developed to deliver real-time heart risk assessment. The project aims to create a trustworthy, transparent, and clinically reliable decision-support tool for cardiovascular healthcare.