Assessment of Heart Patients Dataset Through Regression and Classification Algorithms

This study focused on the survival analysis of patients with heart failure who were admitted to the Institute of Cardiology and Allied Hospital of Faisalabad-Pakistan during April and December 2015. All patients had left ventricular systolic dysfunction, belonging to classes III and IV of the classification carried out by the New York Heart Association. Several Machine Learning algorithms capable of analyzing data through regression and classification techniques were used to predict the mortality rate of future patients with similar problems. Characteristics such as age, ejection fraction, serum creatinine, serum sodium, anemia, platelets, creatinine phosphokinase, blood pressure, diabetes and smoking were considered as potential contributors to mortality. All characteristics were analyzed in order to identify the minimum set of information necessary for a quick and efficient diagnosis of heart failure.