A Rule-Based System for Explainable Donor-Patient Matching in Liver Transplantation

TL;DR

This paper introduces web-liver, a rule-based decision support system for liver donor-patient matching that incorporates causal explanations and aids medical experts in policy design.

Contribution

It presents a novel logic programming interpreter, lppf, extending ASP with partial functions and causal explanations for medical decision support.

Findings

Web-liver effectively models donor-patient matching policies.

The system provides causal explanations for its solutions.

It supports designing policies considering patient severity and transplantation utility.

Abstract

In this paper we present web-liver, a rule-based system for decision support in the medical domain, focusing on its application in a liver transplantation unit for implementing policies for donor-patient matching. The rule-based system is built on top of an interpreter for logic programs with partial functions, called lppf, that extends the paradigm of Answer Set Programming (ASP) adding two main features: (1) the inclusion of partial functions and (2) the computation of causal explanations for the obtained solutions. The final goal of web-liver is assisting the medical experts in the design of new donor-patient matching policies that take into account not only the patient severity but also the transplantation utility. As an example, we illustrate the tool behaviour with a set of rules that implement the utility index called SOFT.