Lebret Audrey
Centre for advanced studies in Biomedical Innovation Law (CeBIL), Copenhagen University, Copenhagen, Denmark.
Paris Human Rights Center (CRDH), Université Paris Panthéon-Assas, Paris, France.
J Law Biosci. 2023 Mar 31;10(1):lsad004. doi: 10.1093/jlb/lsad004. eCollection 2023 Jan-Jun.
Digitization in transplantation is not a new phenomenon. Algorithms are being used, for example, to allocate organs based on medical compatibility and priority criteria. However, digitization is accelerating as computer scientists and physicians increasingly develop and use machine learning (ML) models to obtain better predictions on the chances of a successful transplant. The objective of the article is to shed light on the potential threats to equitable access to organs allocated through algorithms, whether these are the consequence of political choices made upstream of digitization or of the algorithmic design, or are produced by self-learning algorithms. The article shows that achieving equitable access requires an overall vision of the algorithmic development process and that European legal norms only partially contribute to preventing harm and addressing equality in access to organs.
移植领域的数字化并非新现象。例如,算法正被用于根据医学兼容性和优先级标准分配器官。然而,随着计算机科学家和医生越来越多地开发和使用机器学习(ML)模型以更好地预测移植成功的几率,数字化进程正在加速。本文的目的是揭示通过算法分配器官时,在公平获取方面可能面临的潜在威胁,这些威胁是数字化上游的政治选择、算法设计的结果,还是由自学习算法产生的。本文表明,要实现公平获取,需要对算法开发过程有全面的认识,而且欧洲法律规范在预防损害和解决器官获取平等问题方面仅起到部分作用。