人工智能模型改善了慢性肾脏疾病进展到末期肾脏疾病的预测

慢性肾脏疾病（CKD）是一种复杂的疾病，标志着肾功能逐渐下降，最终可以发展为终阶段肾脏疾病（ESRD）。Globally, the prevalence of CKD ranges from 8–16%, with about 5–10% of those diagnosed eventually reaching ESRD, making it a major public health challenge.

在一项新研究中，研究人员使用了机器学习和深度学习模型, as well as explainable artificial intelligence (AI), to assess integrated clinical and claims data with the goal of improving prediction of CKD's progression to ESRD.The integrated models outperformed single data source models, which can enhance CKD management, support targeted interventions, and reduce health care disparities.

研究，卡内基·梅隆大学的研究人员出现在美国医学信息学协会杂志。

"Our study presents a robust framework for predicting ESRD outcomes, improving clinical decision-making through integrated multisourced data and advanced analytics," explains Rema Padman, professor of management science and health care informatics at Carnegie Mellon's Heinz College, who led the study.

“未来的研究将扩大数据整合，并将此框架扩展到其他慢性疾病。”

CKD的进展分为五个阶段，当肾功能降至正常容量的10％至15％，需要透析或移植患者生存。CKD的经济影响很大，美国Medicare CKD患者比例相对较小，造成了不成比例的医疗保险费用，尤其是当他们发展到ESRD时。

In addition, more than a third of ESRD patients are readmitted within 30 days of discharge, underscoring the critical need for early detection and management of the disease to prevent its progression to ESRD, improve patient health outcomes, and reduce health care costs.

In this study, researchers used data from more than 10,000 CKD patients, combining clinical and claims information from 2009 to 2018. They evaluated multiple statistical,机器学习和深度学习模型，使用五个不同的观察窗口。他们的工作得到了可解释的AI的支持，以增强可解释性和降低偏见。

该研究的集成数据模型的表现优于单个数据源模型。一个24个月的观测窗口最佳平衡的早期检测和预测准确性。2021年估计的肾小球滤过率方程提高了预测准确性并降低种族偏见，特别是对于非裔美国人患者。

"Our work bridges a critical gap by developing a framework that uses integrated clinical and claims data rather than isolated data sources," notes Yubo Li, a Ph.D.卡内基·梅隆（Carnegie Mellon）的亨氏学院（Heinz College）的学生，他与该研究合着。

“通过最小化准确预测所需的观察窗口，我们的方法平衡了临床相关性与以患者为中心的实用性；这种整合提高了预测的准确性和临床实用性，从而实现了更明智的决策来改善患者的结果。”

在研究的局限性中，作者说，他们对一个机构的数据的依赖可能会限制其模型对其他护理环境的普遍性。

此外，他们使用数据电子健康记录can introduce observational bias, incomplete records, and underrepresentation of certain patient groups, which can undermine both accuracy and fairness.