开发和验证用于危重患者连续急性肾损伤预测的双通道深度学习
Development and validation of a dual-channel deep learning for continuous acute kidney injury prediction in critically ill patients.
摘要(中文译文)
本研究旨在开发并外部验证一种双通道深度学习模型（DC-AKI），用于对危重症患者48小时急性肾损伤（AKI）风险进行实时连续预测。该模型基于贝斯以色列女执事医疗中心28,099例患者的电子健康记录进行开发，并在两个独立队列中进行外部验证：来自eICU数据库的3,108例患者和来自浙江省人民医院的2,808例患者。31个时间变化特征每6小时更新一次。DC-AKI模型的双通道架构整合了双向门控循环单元（BiGRU）网络、卷积层和注意力机制，以捕捉多尺度时间依赖关系。该模型在内部验证中受试者工作特征曲线下面积（AUC）为0.720（95% CI，0.714-0.728），在两个外部队列中分别为0.577（95% CI，0.570-0.583）和0.798（95% CI，0.795-0.799）。可解释性分析SHAP识别了关键临床预测因子和个体风险轨迹。总之，DC-AKI在开发队列和一个外部验证中心表现出较强的预测性能，但不同机构间的性能差异较大。仍需进一步验证和本地校准以支持其临床部署。
Abstract
This study aimed to develop and externally validate a real-time, continuous prediction model for 48-h acute kidney injury (AKI) risk in critically ill patients using a dual-channel deep learning model (DC-AKI). The model was developed using electronic health records from 28,099 patients at Beth Israel Deaconess Medical Center and externally validated on two independent cohorts: 3,108 patients from the eICU Database and 2,808 patients from Zhejiang Provincial People's Hospital. Thirty-one time-varying features were updated every 6 h. The DC-AKI model's dual-channel architecture integrated BiGRU networks, convolutional layers, and attention mechanisms to capture multiscale temporal dependencies. The model achieved areas under the receiver operating characteristic curve (AUC) of 0.720 (95% CI, 0.714-0.728) in internal validation, and 0.577 (95% CI, 0.570-0.583) and 0.798 (95% CI, 0.795-0.799) in the two external cohorts. Interpretability analysis SHAP identified key clinical predictors and individual risk trajectories. In conclusion, DC-AKI demonstrated strong predictive performance in the development cohort and one external validation site, although performance varied substantially across institutions. Further validation and local calibration are warranted to support its clinical deployment.
如何引用
AMA
Siyu Tang, Chaoqun Niu, Huan Jiang, Yunchao Ling, Zihao Zheng, Jingquan Liu, et al. Development and validation of a dual-channel deep learning for continuous acute kidney injury prediction in critically ill patients.. Renal failure. 2026; doi:10.1080/0886022X.2026.2690700.
APA
Siyu Tang, Chaoqun Niu, Huan Jiang, Yunchao Ling, Zihao Zheng, Jingquan Liu, et al (2026). Development and validation of a dual-channel deep learning for continuous acute kidney injury prediction in critically ill patients.. Renal failure. https://doi.org/10.1080/0886022X.2026.2690700
GB/T 7714
Siyu Tang, Chaoqun Niu, Huan Jiang, Yunchao Ling, Zihao Zheng, Jingquan Liu, et al. Development and validation of a dual-channel deep learning for continuous acute kidney injury prediction in critically ill patients.[J]. Renal failure, 2026 doi:10.1080/0886022X.2026.2690700.
在 PubMed增强版 阅读全文
翻译全文 · AI 文献问答 · PDF 原文下载 · 引用导出