Renal Department, University Hospitals of Coventry and Warwickshire, Coventry, UK.
Research Centre for Health and Life Sciences, Coventry University, Coventry, UK.
Nephrol Dial Transplant. 2024 Nov 27;39(12):2088-2099. doi: 10.1093/ndt/gfae088.
Outcome prediction for live-donor kidney transplantation improves clinical and patient decisions and donor selection. However, the currently used models are of limited discriminative or calibration power and there is a critical need to improve the selection process. We aimed to assess the value of various artificial intelligence (AI) algorithms to improve the risk stratification index.
We evaluated pre-transplant variables among 66 914 live-donor kidney transplants (performed between 1 December 2007 and 1 June 2021) from the United Network of Organ Sharing database, randomized into training (80%) and test (20%) sets. The primary outcome measure was death-censored graft survival. We tested four machine learning models for discrimination [time-dependent concordance index (CTD) and area under the receiver operating characteristic curve (AUC)] and calibration [integrated Brier score (IBS)]. We used decision-curve analysis to assess the potential clinical utility.
Among the models, the deep Cox mixture model showed the best discriminative performance (AUC = 0.70, 0.68 and 0.68 at 5, 10 and 13 years post-transplant, respectively). CTD reached 0.70, 0.67 and 0.66 at 5, 10 and 13 years post-transplant. The IBS score was 0.09, indicating good calibration. In comparison, applying the Living Kidney Donor Profile Index (LKDPI) on the same cohort produced a CTD of 0.56 and an AUC of 0.55-0.58 only. Decision-curve analysis showed an additional net benefit compared with the LKDPI 'treat all' and 'treat none' approaches.
Our AI-based deep Cox mixture model, termed Live-Donor Kidney Transplant Outcome Prediction, outperforms existing prediction models, including the LKDPI, with the potential to improve decisions for optimum live-donor selection by ranking potential transplant pairs based on graft survival. This model could be adopted to improve the outcomes of paired exchange programs.
活体供肾移植的预后预测可以改善临床和患者决策以及供者选择。然而，目前使用的模型的区分度或校准能力有限，因此迫切需要改进选择过程。我们旨在评估各种人工智能（AI）算法在改善风险分层指数方面的价值。
我们评估了来自美国器官共享网络数据库的 66914 例活体供肾移植（2007 年 12 月 1 日至 2021 年 6 月 1 日期间进行）的移植前变量，将其随机分为训练（80%）和测试（20%）集。主要观察终点为死亡风险校正移植物存活率。我们测试了四种机器学习模型的区分度[时间依赖性一致性指数（CTD）和接受者操作特征曲线下面积（AUC）]和校准[综合 Brier 评分（IBS）]。我们使用决策曲线分析评估潜在的临床实用性。
在这些模型中，深度 Cox 混合模型显示出最佳的区分性能（移植后 5、10 和 13 年的 AUC 分别为 0.70、0.68 和 0.68）。CTD 分别为 0.70、0.67 和 0.66，在移植后 5、10 和 13 年。IBS 评分是 0.09，表明校准良好。相比之下，在同一队列中应用活体供肾供者特征指数（LKDPI）仅产生 CTD 为 0.56 和 AUC 为 0.55-0.58。决策曲线分析显示与 LKDPI“治疗所有”和“不治疗任何”方法相比，具有额外的净获益。
我们基于 AI 的深度 Cox 混合模型，称为活体供肾移植预后预测，优于现有的预测模型，包括 LKDPI，具有通过基于移植物存活率对潜在移植对进行排名来优化活体供者选择的潜力。该模型可以用于改善配对交换计划的结果。
