Zhang Li, Li Mingyu, Wang Chengcheng, Zhang Chi, Wu Hong
School of Medicine and Health Management, Huazhong University of Science and Technology, Wuhan, China.
Yunnan Provincial Archives, Kunming, China.
Digit Health. 2025 Jan 6;11:20552076241311173. doi: 10.1177/20552076241311173. eCollection 2025 Jan-Dec.
Acute kidney injury (AKI) poses a lethal risk in intensive care unit (ICU) patients, where early detection is challenging. This study was to establish a prediction model for AKI 24 hours in advance for ICU patients and to help clinicians monitor patients at an early stage by key features.
In this study, the Medical Information Mart for Intensive Care (MIMIC) databases were used to construct a dataset of critically ill patients. Predictive models were constructed using five machine learning algorithms based on MIMIC-IV data, and the best predictive model was selected by multiple model evaluation metrics. MIMIC-III data were used for external validation. We conducted an interpretability analysis of the model using SHapley Additive exPlanations (SHAP) to clarify key features and decision-making mechanisms.
A total of 18,186 patient data were included in this study. The analysis combining calibration and decision curves demonstrated that the eXtreme Gradient Boosting (XGBoost) exhibited superior performance among the five algorithms, achieving an area under the receiver operating characteristic curve of 0.88. Interpretability analysis based on the XGBoost model showed diuretic use, mechanical ventilation, vasopressor use, age, and antibiotic use as the most important decision factors of the model. The SHAP summary plot was used to illustrate the effects of the top 19 features attributed to the XGBoost.
The XGBoost algorithm can predict the occurrence of AKI more accurately. Interpretative analysis of the model reveals the mechanisms of key features, and reflects the individual differences between patients, providing an important clinical reference.
急性肾损伤（AKI）对重症监护病房（ICU）患者构成致命风险，早期检测具有挑战性。本研究旨在建立一个提前24小时预测ICU患者发生AKI的模型，并通过关键特征帮助临床医生进行早期监测。
在本研究中，使用重症监护医学信息集市（MIMIC）数据库构建危重病患者数据集。基于MIMIC-IV数据，使用五种机器学习算法构建预测模型，并通过多种模型评估指标选择最佳预测模型。MIMIC-III数据用于外部验证。我们使用SHapley加性解释（SHAP）对模型进行可解释性分析，以阐明关键特征和决策机制。
本研究共纳入18186例患者数据。校准和决策曲线相结合的分析表明，极端梯度提升（XGBoost）在五种算法中表现出卓越性能，受试者操作特征曲线下面积达到0.88。基于XGBoost模型的可解释性分析显示，利尿剂使用、机械通气、血管活性药物使用、年龄和抗生素使用是该模型最重要的决策因素。使用SHAP汇总图来说明归因于XGBoost的前19个特征的影响。
XGBoost算法能够更准确地预测AKI的发生。对该模型的解释性分析揭示了关键特征的机制，反映了患者之间的个体差异，提供了重要的临床参考。
