Prediction of Different Symptoms Emergence in Heart Failure Patients Using Machine Learning Approaches

Heart failure presents with a wide range of symptoms that affect patients' quality of life. This study uses physical-examination data and blood biomarkers to predict the emergence of thirteen individual HF-related symptoms (e.g. dyspnea, orthopnea, peripheral oedema, pulmonary crackles) as present/absent binary outcomes. Using Extreme Gradient Boosting, the models correctly predicted symptom presence in 88.79% of cases on average, ranging from 74.16% (dyspnea at rest/activity) to 98.47% (pretibial edema) accuracy, suggesting symptom emergence can be anticipated from HF severity and biomarker values. This work was presented at the 4th Serbian International Conference on Applied Artificial Intelligence (SICAAI 2025), Zlatibor, Serbia, and was carried out within the STRATIFYHF project.