Predicting Heart Failure Survival: A Machine Learning Approach with Explainable AI

Heart failure is one of the global health issues that is needed for precise prognostic instruments. In order to predict survival and time-to-event outcomes in patients with heart failure, this study assesses machine learning techniques. Most effective prediction models are determined for heart failure prognosis by combining clinical information and evaluating model performance. According to the results, the LightGBM classifier had the lowest accuracy at 77%, while the SVM classifier had the highest accuracy at 88%. These findings suggest that machine learning can greatly improve clinical judgement in cardiovascular medicine and advance patient-focused treatment.