Heart Failure Survival Prediction Model

Document Type

Abstract

Publication Date

Spring 5-1-2025

Abstract

Predicting patient survival rates in heart failure cases not only optimizes treatment strategies and improves patient care but also provides families with a way to estimate survival probabilities. This helps them navigate the uncertainty surrounding their loved one's health condition. By offering a reliable prediction, this prediction system assists families in making informed decisions and preparing for potential outcomes, ultimately reducing stress and emotional burden. Additionally, it enhances clinical decision-making for healthcare professionals, enabling more effective risk assessment, personalized treatment planning, and resource allocation in managing heart failure patients. (Maziar Sabouri et al.) This project develops a machine learning-based Heart Failure Survival Prediction Model, which predicts a patient's likelihood of survival based on clinical data. The model is trained using various patient attributes, including age, gender, smoking, and diabetes. (Moreno-Sánchez) The proposed system utilizes multiple machine learning approaches to ensure accurate predictions. If all 14 clinical variables are available, the model automatically selects the most optimal prediction algorithm based on performance metrics. However, if only a subset of variables (n variables) is provided, a Random Forest model is used, as it can handle missing features effectively. By implementing this solution, healthcare professionals make data-driven decisions to better assess and manage the risks associated with heart failure patients. (Tang and Ishwaran).

Comments

Completed as part of the Computer Science Senior Capstone Project.

This document is currently not available here.

Share

COinS