Research on Predicting Survival in Heart Failure Patients of Logistic Regression Models

Abstract

Abstract. The purpose of this paper is to study a logistic regression model for predicting the survival of patients with heart failure. Heart failure is a serious clinical syndrome that usually develops from multiple heart diseases, and its inadequate pumping function of the heart poses a significant threat to the life and health of the patient and a significant economic burden on the healthcare system. Using medical record data from 299 patients with heart failure in the UCI database, this research utilized logistic regression models to identify the critical factors influencing the survival of heart failure patients and to forecast their survival outcomes. This paper found that age, ejection fraction, serum creatinine concentration and follow-up period are significant factors affecting survival in patients with heart failure. By adopting backward elimination method to optimize the model, the accuracy of prediction is further improved. The optimized logistic regression model yields an area under the Receiver Operating Characteristic (ROC) curve of 0.845, showing high prediction accuracy. The conclusions of this study provide a new perspective for the early diagnosis, risk assessment and personalized treatment of heart failure.