Title: Predicting Heart Failure Mortality Using Artificial Neural Networks: A Clinical Data-Driven Approach
Authors: Basheer Ali Ghazal, Abdullah Salam Alnkhala, Samy S. Abu-Naser
Volume: 10
Issue: 7
Pages: 14-18
Publication Date: 2026/07/28
Abstract:
Heart failure (HF) remains a leading cause of mortality worldwide, necessitating early and accurate risk stratification to improve patient outcomes. This study develops a predictive model for heart failure mortality utilizing Artificial Neural Networks (ANNs) applied to a clinical dataset of 299 patients. The dataset encompasses 13 clinical features, including age, serum creatinine, ejection fraction, and comorbidities such as diabetes and hypertension. Drawing upon the structural and algorithmic foundations outlined in "Elements of Artificial Neural Networks", a Multi-Layer Perceptron (MLP) architecture trained via Backpropagation was implemented. The methodology involves preprocessing, feature importance extraction, and network optimization. The developed neural network demonstrated robust performance in predicting death events, with serum creatinine levels and ejection fraction emerging as the most significant clinical predictors. The findings demonstrate that integrating ANN models into clinical decision support systems can substantially enhance prognosis accuracy and optimize patient care protocols.