Mortality Prediction in Heart Failure Patients Using Machine Learning Techniques
Abstract
Heart failure is a critical cardiovascular condition affecting millions globally. Early prediction of mortality risk among heart failure patients can help guide treatment strategies and potentially save lives. In this study, we use the Heart Failure Clinical Records Dataset to develop predictive models using machine learning algorithms. We applied Logistic Regression, Random Forest, and Support Vector Machine (SVM) classifiers to predict the likelihood of patient death. Our results indicate that the Random Forest model achieved the highest accuracy (91%), suggesting its effectiveness in handling medical datasets with mixed feature types. The study supports the use of predictive analytics to aid clinical decision-making.
Keywords
Download Options
Introduction
Cardiovascular diseases (CVDs) are the leading cause of death worldwide, with heart failure being one of the major contributors. Heart failure occurs when the heart cannot pump sufficient blood to meet the body's needs. Accurate and early prediction of mortality risk in heart failure patients is critical for clinical decision-making and resource allocation.
The goal of this research is to build predictive models to classify patients into two categories: those who are at risk of death and those who are not. We employ supervised machine learning algorithms to create models based on patient clinical records, helping clinicians make informed decisions.
Conclusion
The study confirms the efficacy of machine learning algorithms in predicting heart failure mortality. Random Forest outperformed both Logistic Regression and SVM, indicating its robustness for clinical datasets. The implementation of such models in hospital decision-support systems can assist doctors in identifying high-risk patients early. Future work should explore deep learning approaches and cross-institutional validation to enhance model generalizability.