Understanding and Predicting Organ Donation Outcomes Using Network-based Predictive Analytics

Abstract

There is an ever-increasing disparity between the number of organs needed for transplantation and the number available for donation. As a result, thousands of people die every year while waiting for an organ transplant. Therefore, it is now more critical than ever to study the factors associated with organ donation. A better understanding of such factors will help immeasurably in formulating data-driven strategies for improving familial consent for organ donation. This research combines machine learning methods and network science to accurately predict organ donation consent outcomes. In this study, six years of patient data from an organ procurement organization (OPO) in New York City were obtained and used to propose the consent prediction model. A comparison of the various prediction models was also conducted. OPOs can now use the best models to develop strategies for optimizing the consent rate, thereby saving more lives. The experimental results show that our approach outperformed in terms of detection because we combined network and machine learning algorithms to obtain clearer insights. The proposed approach can be used as an expert system to increase the organ donation consent rate, thereby bridging the gap between organ demand and supply.