Utilizing Machine Learning to Predict Heart Transplant Allocation and Evaluate Gender Disparities

Overview

For patients with end-stage heart failure, receiving a heart transplant (HTx) increases their one-year survival chances by nearly fourfold. With 20% experiencing depression and 48.9% waiting over a year for a transplant, reducing uncertainty and ensuring allocation is equitable is critical. Unlike kidney and liver transplantation, no publicly available HTx prediction tool exists. Secondly, although revised 2018 United Network for Organ Sharing (UNOS) policies improved overall transplant access for women, gender disparities persist today among the most critically ill (Status 1) patients. This research develops a Machine Learning based HTx prediction tool and uncovers why Status 1 women experience lower rates of transplantation. UNOS data of adult HTx patients listed between September 2019 to July 2025 was analyzed. A feedforward multilayered perceptron (MLP) model containing 20+ predictor variables successfully predicted total days on the waiting list with a mean absolute error of 11.7 days. To investigate gender disparities, two MLP multi-class classifiers were trained to predict allocation for female and male candidates. Model weight analysis revealed distinct gender-based predictive factors: for women, physical limitations, such as height and limited ventricular assist device compatibility, were barriers to transplantation; men were influenced positively by socioeconomic factors like highest level of education. This research demonstrates the ability to provide an accurate predictive model for HTx candidates and identifies the key biological and structural factors underlying gender-based allocation disparities amongst the high-priority Status 1 patients.

Competition history

• CSEF 2026 Medicine & Physiology (Senior Division) · Entry S-15-35

Related projects

Closest projects by meaning, across every fair and year in the corpus.

Browse more like this

• All Biomedical and Health Sciences projects
• CSEF Biomedical and Health Sciences projects
• Every CSEF 2026 project