Maryland Smith’s Michael Fu Works to Improve Kidney Transplants With NSF Grant

Maryland Smith’s Michael Fu received a National Science Foundation grant for research to improve kidney transplants. The project combines technological advances in medical science with operations research and artificial intelligence-based approaches to dramatically improve the kidney donor-recipient matching process to lead to better outcomes for end-stage kidney disease patients.

Fu is working with researchers at George Mason University and the University of Louisville. The collaborative project, “Optimal Desensitization Protocol in Support of a Kidney Paired Donation (KPD) System,” is supported by nearly $1 million in three NSF grants.

The goal of the research is to increase the number of successful kidney transplants, decrease waiting times for donor organs, and lower costs for healthcare providers and patients. The researchers say kidney transplants are well-suited to data-driven approaches, allowing Fu and his fellow researchers to build better simulation models and optimize data-driven algorithms for a potentially far greater impact.

In contrast to other major organs (e.g., heart, lung), a large proportion of kidney transplants come from living donors. To increase the transplant pool, the medical community uses a kidney paired donation system/exchange that allows transplant candidates to swap their incompatible living donor’s kidney for a compatible one. Another medical development for increasing the likelihood of an organ transplant match is to use “desensitization” therapy to remove antibodies from a transplant recipient’s bloodstream to allow acceptance of otherwise incompatible organs.

Fu’s NSF grant – for $159,904 – will mainly cover his work on the patient behavior model, which will incorporate complex uncertainties and risks, including the option of desensitization therapy. It is envisioned that this model will lead to a decision-support tool to help patients determine whether to accept an offered kidney that may require desensitization therapy, or to risk waiting for a more compatible one at some unknown point in the future. Currently, there is no tool that assists patients with the decision whether to accept an offered kidney, even without the desensitization option. This model will be part of a larger framework for data-driven decision-making to match kidney donors and recipients using simulation optimization.

Fu is the Smith Chair of Management Science in the decisions, operations and information technologies department. He has a joint appointment with the Institute for Systems Research and an affiliate appointment with the Department of Electrical and Computer Engineering, both in the A. James Clark School of Engineering.