Multi-Stage Classification Approach for Heart Failure Disease Diagnosis and Reduced Ejection Fraction Prediction

Heart failure (HF) is one of the most common medical conditions worldwide. Left ventricular ejection fraction (LVEF) is one of the most important features used to further classify HF patients into patients with reduced ejection fraction (HFrEF), mid-range ejection fraction (HFmrEF) and preserved ejection fraction (HFpEF). This paper describes a machine learning based pipeline for multi-stage classification of patients according to presence of HF and degree of reduced ejection fraction, using echocardiography, physical examination and demographic data collected within the STRATIFYHF project. Two separate pipelines were created: the first classifies HF patients into 3 classes based on ejection fraction (reduced, mid-range and preserved); the second conducts classification into LVEF<50% and LVEF>50%, followed by a second classification of HFrEF and HFpEF classes. The HF classification model achieved 97% accuracy and 98% F1 score for the confirmed HF class. The final two-stage classification model achieved an overall accuracy of 96% and 87% for LVEF>50% versus LVEF<50% and mid-range versus reduced ejection fraction respectively. Presented at the 25th IEEE International Conference on Bioinformatics and Bioengineering (BIBE 2025), 6-8 November 2025, Athens, Greece.