Enterprise AI Analysis

Unlocking Transparent & Accurate Heart Disease Prediction with XAI-HD

Cardiovascular disease (CVD) remains the leading global cause of death, emphasizing the critical need for advanced diagnostic tools. Traditional AI models often lack transparency, hindering clinical adoption. This study introduces XAI-HD, a hybrid framework that integrates machine learning (ML), deep learning (DL), and Explainable AI (XAI) techniques for robust and interpretable heart disease detection. XAI-HD systematically addresses challenges like class imbalance, missing data, and feature inconsistency through advanced preprocessing and class-balancing methods (OSS, NCR, SMOTEN, ADASYN, SMOTETomek, SMOTEENN). Comparative evaluations across multiple datasets (CHD, FHD, SHD) demonstrate a significant reduction in classification error rates by 20–25% compared to traditional ML models, achieving superior accuracy, precision, recall, and F1-score. Crucially, SHAP and LIME-based feature importance analysis enhances model interpretability, fostering trust among medical professionals. The framework is designed for seamless integration into hospital decision support systems and real-time cardiac risk assessment, offering a balanced, interpretable, and computationally efficient solution for clinical environments.