Analysis of Online Algorithms for Organ Allocation.
Abstract
In this paper we discuss the need for a refinement to the existing organ allocation policy. Under the current organ allocation algorithm - the Point System - whenever an organ becomes available each patient scores points for that organ according to a multitude of criteria, and the patient with the highest score gets the organ. We will show that in order to select the patient that should receive the organ we must look not only at each patient individually but also at the current collection of patients. Current work in progress shows that the expected distribution of patients as well as the expected distribution of organs should affect the allocation policy. We suggest an objective function for a case when there are no overwhelming considerations to prefer one patient over the another (i.e. they have the same score, or are very close). The objective function we analyze is the number of organs utilized. The reason is that organs are considered to be a scarce resource not to be wasted. We will select a patient in such a way that as few possible organs will be wasted in the future. In this paper we will solve, using the online model, some simplifications of the general problem. We will explain the intuition that we glean from them. We will also present some preliminary results from analyzing the problem using statistical and simulation methods.
Document Details
- Document Type
- Technical Report
- Publication Date
- Oct 03, 1990
- Accession Number
- ADA249361
Entities
People
- Shmuel Ur
Organizations
- Carnegie Mellon University