Collaborative Research: FW-HTF-R: Embedding Preferences in Adaptable Artificial Intelligence Decision Support for Transplant Healthcare to Reduce Kidney Discard

与慢性透析相比，移植为终末期肾病患者提供了更好的生活质量和长期生存。然而，大约20%的死亡供体肾脏被丢弃，从未移植。虽然有些丢弃可能是医学上适当的，但其他人反映了错过的机会。即使是被认为不太理想的肾脏也可能为一些患者提供生存益处。器官采购组织（OPO）很难找到移植中心接受医学上不太理想（风险较高）的肾脏。根据他们的判断，OPO可以使用加速放置来绕过“难以放置”肾脏的优先列表。然而，由于缺乏数据驱动的指导，这一机制没有得到系统应用，而且很可能没有得到充分利用。为了实现变革，我们将把人工智能（AI）决策支持整合到肾脏供应过程中，以满足移植中心的需求和OPO的供应。主要工作人员包括OPO工作人员（器官采购协调员，手术主任，医疗主任），移植中心工作人员（协调员，医生，外科医生）和移植患者。这项研究是由圣刘易斯大学医院的移植和伦理专家、联合器官共享网络（UNOS）的行为科学家以及密苏里州科技大学的人工智能和人为因素专家合作推动&的。该项目正在开发一个人工智能决策支持系统，用于（a）移植中心接受/拒绝高风险肾脏供应，以及（B）OPO更快地识别难以放置的肾脏。这项研究将（1）测量工人偏好以定制支持系统的操作和界面，（2）聚合不同利益相关者定义的公平偏好以提高模型输出的公平性，（3）评估将不确定性和可解释性嵌入界面的效果，（4）开发深度学习集成模型，该模型可以随着时间的推移而适应，同时可以解释，以及（5）使用UNOS实验室的SimUNet进行随机对照试验，这是一个用于行为实验的真实肾脏提供模拟平台，以估计对肾脏丢弃的影响。在深度学习模型中，该项目将在不显著降低准确性的情况下进行权衡，以提高公平性，通过将特征相关性转换为语言表达来增强可解释性，并通过迁移学习来整合新数据（例如根据工人偏好进行定制）。最终，这项研究旨在将“难以放置”器官的肾脏丢弃减少至少10%。此外，这项工作将支持道德和培训方面的关键进展，这些问题对于克服将人工智能融入医疗保健的系统级障碍至关重要。该奖项反映了NSF的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

Transplantation provides patients suffering from end-stage kidney disease a better quality of life and long-term survival compared to chronic dialysis. However, approximately 20% of deceased donor kidneys are discarded and never transplanted. While some discards may be medically appropriate, others reflect missed opportunities. Even kidneys deemed less desirable may provide survival benefits to some patients. Organ Procurement Organizations (OPOs) have great difficulty finding transplant centers to accept less medically desirable (higher risk) kidneys. At their discretion, OPOs can use accelerated placement to bypass the priority list for “hard-to-place” kidneys. However, due to a lack of data-driven guidance, this mechanism is not systematically applied and likely underutilized. To enable transformative change, we will integrate Artificial Intelligence (AI) decision support into the kidney offer process for both demand at the transplant center and supply at the OPO. Key workers include OPO staff (organ procurement coordinators, operations directors, medical directors), transplant center staff (coordinators, physicians, surgeons), and transplant patients. This research is driven by a partnership between transplant and ethics experts at Saint Louis University Hospital, behavioral scientists at the United Network for Organ Sharing (UNOS), and experts in AI and human factors from Missouri University of Science & Technology.Building on a FW-HTF planning grant, this project is developing an AI decision support system for (a) transplant centers to accept/deny high-risk kidney offers and (b) OPOs to identify hard-to-place kidneys sooner. This research will (1) measure worker preferences to customize the support system’s operation and interface, (2) aggregate fairness preferences as defined by diverse stakeholders to improve fairness in the model output, (3) evaluate the effect of embedding uncertainty and explainability into the interface, (4) develop deep learning ensemble models that can adapt over time while being explainable, and (5) conduct randomized control trials using UNOS Lab’s SimUNet, a realistic kidney offer simulation platform for behavioral experiments, to estimate the impact on kidney discard. Within the deep learning model, this project will impose trade-offs to increase fairness without significantly reducing accuracy, enhance explainability by converting feature relevance into linguistic expressions, and integrate new data (such as customizing for worker preferences) through transfer learning as conditions change in kidney transplant practices. Ultimately, this research aims to reduce kidney discard for “hard-to-place” organs by at least 10%. In addition, this work will support critical advancements in ethics and training, issues that will be critical in overcoming system-level barriers to integrate AI into healthcare.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.