A data science approach for quantifying spatio-temporal effects to graft failures in organ transplantation

Abstract

The transplantation of solid organs is one of the most important accomplishments of modern medicine. Yet, organ shortage is a major public health issue; 8,000 people died while waiting for an organ in 2014. Meanwhile, the allocation system currently implemented can lead to organs being discarded and the medical community still investigates factors that affects early graft failure such as distance and ischemic time. In this paper, we investigate early graft failure under a spatio-temporal perspective using a data science unified approach for all six organs that is based on complementary cumulative analysis of both distance and ischemic time. Interestingly, although distance seems to highly affect some organs (e.g. liver), it appears to have no effect on others (e.g. kidney). Similarly, the results on ischemic time confirm it affects early graft failure with higher influence for some organs such as (e.g. heart) and lower influence for others such as (e.g. kidney). This poses the question whether the allocation policies should be individually designed for each organ in order to account for their particularities as shown in this work.