Discrete-Survival Compatibility Score for Donor-Recipient Matching in Liver Transplantation

Le résumé fourni par la source

Liver transplantation is a life-saving treatment for patients with end-stage liver disease. However, donor organ scarcity and patient heterogeneity make optimal donor recipient matching a persistent challenge. In this work, we present a two-step approach that models post-transplant survival probabilities at clinically relevant discrete time points and combines the predictions into a compatibility score for each donor-recipient pair. Our findings suggest that, for the multi-center liver transplant dataset, our approach yields higher accuracy than directly estimating a single compatibility score. By integrating statistical and machine learning techniques into a cohesive framework, empirical results show that our method outperforms traditional methods while preserving clinical interpretability, which enables more nuanced donor-recipient matching and supports practical decision-making in real-world clinical settings.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.