Advanced Machine Learning Techniques for Accurate Heart Disease Prediction

As one of the leading causes of death worldwide, heart disease necessitates early detection and treatment to improve patient outcomes and precise prognosis. The performance of eight sophisticated machine learning approaches is assessed and contrasted in this project - CatBoost, LightGBM, AdaBoost with SMOTE, Transformer-Based Models, Explainable Boosting Machine (EBM), and CNN-LSTM Models—applied to a heart disease prediction dataset obtained from Kaggle. Key evaluation metrics like the following are accustomed to assess the models: accuracy precision F1-score, recall and AUC-ROC to identify the most effective approach for heart disease prediction. It is observed that LightGBM acquired 91.2% accuracy compared to other models.