Learning Matching Representations for Individualized Organ Transplantation Allocation

Organ transplantation is often the last resort for treating end-stage illness, but the probability of a successful transplantation depends greatly on compatibility between donors and recipients. Current medical practice relies on coarse rules for donor-recipient matching, but is short of domain knowledge regarding the complex factors underlying organ compatibility. In this paper, we formulate the problem of learning data-driven rules for organ matching using observational data for organ allocations and transplant outcomes. This problem departs from the standard supervised learning setup in that it involves matching the two feature spaces (i.e., donors and recipients), and requires estimating transplant outcomes under counterfactual matches not observed in the data. To address these problems, we propose a model based on representation learning to predict donor-recipient compatibility; our model learns representations that cluster donor features, and applies donor-invariant transformations to recipient features to predict outcomes for a given donor-recipient feature instance. Experiments on semi-synthetic and real-world datasets show that our model outperforms state-of-art allocation methods and policies executed by human experts. Proceedings of the 24 International Conference on Artificial Intelligence and Statistics (AISTATS) 2021, San Diego, California, USA. PMLR: Volume 130. Copyright 2021 by the author(s). Observational training data for Organ allocation Matching decisions for new patients Good generalization performance Poor generalization performance Figure 1: Donor-recipient matching for organ transplantation. We show an exemplary training data set with two types of donors and recipients (red and blue in either case). In current practice, blue recipients are consistently allocated red organs, and vice versa; hence we only observe blue-red and red-blue matches in training data. Using a supervised learning model f to predict transplant outcomes for alternative allocations would provide accurate predictions for blue-red/red-blue matches but would generalize poorly to blue-blue/red-red matches, making it challenging to learn new allocation rules from the training data.