Date of Award

January 2026

Document Type

Open Access Thesis

Degree Name

Medical Doctor (MD)

Department

Medicine

First Advisor

F. P. Wilson

Abstract

End-stage renal disease (ESRD) after heart transplantation is a common complicationwhich increases risk of mortality. Simultaneous heart-kidney transplantation (SHKT) lowers this risk. Eligibility for SHKT relies on pre-transplant estimated glomerular filtration rate (eGFR) with ongoing discussion regarding optimal eligibility guidelines. We present an eXtreme Gradient Boosting Decision Tree (XGBoost) machine learning model for predicting ESRD within one year of heart transplant procedure in candidates ineligible for kidney-alone transplantation. The model was trained using readily available elements from pre-transplant evaluation and performed with an area under the receiver operating characteristic curve (AUROC) of 0.71. The model discrimination was statistically significantly better than a pre-transplant eGFR based logistic regression model which represents current standard of care. Decision curve analysis demonstrated net clinical benefit from model utilization at the optimal threshold probability.

Comments

This is an Open Access Thesis.

Download
Open Access

This Article is Open Access

Share

COinS