Heart Failure Prediction Using Support Vector Machine

Heart failure is a significant global health challenge, requiring effective and early diagnostic tools to improve patient outcomes. In this study, we developed a predictive model for heart failure using Support Vector Machines (SVM), leveraging clinical data from 299 patients. The dataset includes key features such as age, ejection fraction, serum sodium levels, and comorbidities like diabetes and high blood pressure. Our SVM model demonstrated exceptional predictive performance, achieving a training accuracy of 99.7% and a testing accuracy of 99.1%.