Master Project in Collaboration with Agence de Biomédecine

Please find below the description of a Master Project on Kidney Allocation. 

The project will start in February 2017 and will last for 5 to 6 months. 

Location : Paris and/or Grenoble

Please contact gautier.stauffer@grenoble-inp.fr for further information.

-----

Organ Allocation for Kidney Transplantation

1 Context and objective

Transplantation and dialysis are the only treatments for severe forms of kidney diseases. Transplantation is usually a better option for patients since it improves life years survival and quality of life. In France, most of the transplanted organs come from deceased donors: patients join a waiting list and when an organ becomes available, the list is screened by priority until a compatible patient is found. Priority is based on a Kidney Allocation Score (KAS). The KAS of a patient/organ pair typically takes into account: the characteristics of both the patient and the organ (such as ABO blood type, age, HLA, weight, size, etc...), the characteristics of the patient (such as the number of years spent in the waiting list, etc...) and characteristics of the pair (such as the expected post transplant survival of the patient if given the organ). There are many ways of assigning scores to patient/organ pairs. We focus here on KAS defined as weighted sums of “component functions” defined over the various characteristics. One of the main challenges is to design fair and efficient KAS.

2 Approach

“Agence de la Biomédecine” has already implemented a Kidney Allocation Score which is functional. The goal of this project is to evaluate whether operations research techniques can help “Agence de la Biomédecine” in improving their Kidney Allocation Score. We propose to develop an algorithm following the methodology introduced by Bertsimas, Farias and Trichakis [1] to support US policymakers in this same task. Fairness conditions, “component functions” and objective functions will be given as input by “Agence de la Biomédecine” and the main challenge is to design a set of weights on the component functions so as to maximize the objective functions while respecting as much as possible the fairness criterion. The proposed approach is data-driven and uses simple, yet powerful, techniques from operations research and statistical analysis, essentially linear programming and linear regression. In a second phase, we might analyse the impact of fairness conditions and component functions on the objective functions, and propose alternative options that might improve the overall quality of the protocol.

3 Supervision

This project stems from a collaboration between G-SCOP laboratory in Grenoble and “Agence de la Biomédecine” in Saint-Denis La Plaine (93). The student will work together with both parties : “Agence de la Biomédecine” will provide practical expertise on the subject. In particular it will explain the various component functions and objective functions used in the design of the scoring rule, and it will provide historical data on patients and organs ; G-SCOP will lead the optimization part. Supervisors : “Agence de la biomédecine” : Benoit Audry, Christian Jacquelinet, Marie-Alice Macher ; G-SCOP : Lucie Pansart, Gautier Stauffer.

References

[1] Dimitris Bertsimas, Vivek F Farias, and Nikolaos Trichakis. Fairness, efficiency, and flexibility in organ allocation for kidney transplantation. Operations Research, 61(1):73–87, 2013.