Benchmarking Waitlist Mortality Prediction in Heart Transplantation Through Time-to-Event Modeling using New Longitudinal UNOS Dataset
Researchers have developed and benchmarked machine learning models to predict waitlist mortality for heart transplant patients. Using a new longitudinal dataset from the United Network for Organ Sharing (UNOS) with 23,807 patient records, their best model achieved a C-Index of 0.94 and AUROC of 0.89. This performance significantly surpasses previous models and can aid in assessing patient urgency and refining transplant policies. AI
IMPACT Improves predictive accuracy for critical medical decisions, potentially saving lives and optimizing resource allocation in organ transplantation.