EURO 2024 Copenhagen
Abstract Submission

EURO-Online login

2496. Synthetic data generation for kidney exchange programme simulation

Invited abstract in session TD-10: Kidney Exchange II, stream OR in Health Services (ORAHS).

Tuesday, 14:30-16:00
Room: 11 (building: 116)

Authors (first author is the speaker)

1. William Pettersson
University of Glasgow
2. David Manlove
School of Computing Science, University of Glasgow

Abstract

Kidney exchange programmes (KEPs) increase the potential for living kidney donation, often using operational research techniques, and drastically improving the quality of life of recipients. This is achieved by identifying an optimal set of transplants that can be performed within a pool of donor-recipient pairs via exchange cycles and chains. However, determining appropriate optimality criteria to use is not trivial, and only with historical real data or realistic synthetic data can simulations show how different criteria affect the performance of the KEP. However, historical datasets can be difficult to obtain.

In this talk we walk through the process of producing a realistic synthetic data generator from real-world data using state-of-the-art techniques. We include full Jupyter notebooks that can easily be followed by researchers associated with real-world KEPs. We then follow up by demonstrating how the data generator can be used to simulate the effects of a number of different optimality criteria on the KEP.

Keywords

Status: accepted

Back to the list of papers