Heart Failure Prediction: Performance Evaluation and Comparative Analysis of Machine Learning Algorithms

Heart failure remains a leading cause of morbidity and mortality worldwide, highlighting the need for effective predictive models to enhance clinical decision-making. This study aims to conduct a performance analysis and comparative review of the most relevant research utilizing machine learning algorithms for heart failure prediction. We started with a literature exploration phase to select recent studies published between 2020 and 2024. We examined various machine learning methods, including random forest (RF), decision tree (DT), Naive Bayes, support vector machine (SVM), logistic regression, neuro-fuzzy system (NFS), k-modes clustering, AdaBoost, and light gradient boosting machine (GBM). The performance of these models was assessed using key metrics such as accuracy, sensitivity, specificity, and F1-score. Our analysis also included the number of features and the datasets employed in each study. Additionally, we discussed the challenges identified in the literature, providing a comprehensive overview of the current state of machine learning applications in heart failure prediction. The comparative study indicates that performance is heavily influenced by the type of model used, the number of features incorporated, and the specific dataset employed. Ultimately, machine learning models have the potential to aid clinical decision-making by predicting heart failure outcomes with accuracies reaching up to 100% for the NFS technique and 93.26% for the GBM.