Emergency Medicine Clinical Research Center, Beijing Key Laboratory of Cardiopulmonary Cerebral Resuscitation, Beijing Chaoyang Hospital Affiliated to Capital Medical University, Beijing, 100020, China.
Department of Health Management, Shandong Engineering Laboratory of Health Management, Institute of Health Management, the First Affiliated Hospital of Shandong First Medical University & Shandong Provincial Qianfoshan Hospital, Jinan, 250014, China.
BMC Pulm Med. 2023 Oct 3;23(1):370. doi: 10.1186/s12890-023-02663-6.
BACKGROUND: Acute kidney injury (AKI) can make cases of acute respiratory distress syndrome (ARDS) more complex, and the combination of the two can significantly worsen the prognosis. Our objective is to utilize machine learning (ML) techniques to construct models that can promptly identify the risk of AKI in ARDS patients. METHOD: We obtained data regarding ARDS patients from the Medical Information Mart for Intensive Care III (MIMIC-III) and MIMIC-IV databases. Within the MIMIC-III dataset, we developed 11 ML prediction models. By evaluating various metrics, we visualized the importance of its features using Shapley additive explanations (SHAP). We then created a more concise model using fewer variables, and optimized it using hyperparameter optimization (HPO). The model was validated using the MIMIC-IV dataset. RESULT: A total of 928 ARDS patients without AKI were included in the analysis from the MIMIC-III dataset, and among them, 179 (19.3%) developed AKI after admission to the intensive care unit (ICU). In the MIMIC-IV dataset, there were 653 ARDS patients included in the analysis, and among them, 237 (36.3%) developed AKI. A total of 43 features were used to build the model. Among all models, eXtreme gradient boosting (XGBoost) performed the best. We used the top 10 features to build a compact model with an area under the curve (AUC) of 0.850, which improved to an AUC of 0.865 after the HPO. In extra validation set, XGBoost_HPO achieved an AUC of 0.854. The accuracy, sensitivity, specificity, positive prediction value (PPV), negative prediction value (NPV), and F1 score of the XGBoost_HPO model on the test set are 0.865, 0.813, 0.877, 0.578, 0.957 and 0.675, respectively. On extra validation set, they are 0.724, 0.789, 0.688, 0.590, 0.851, and 0.675, respectively. CONCLUSION: ML algorithms, especially XGBoost, are reliable for predicting AKI in ARDS patients. The compact model maintains excellent predictive ability, and the web-based calculator improves clinical convenience. This provides valuable guidance in identifying AKI in ARDS, leading to improved patient outcomes.
背景：急性肾损伤 (AKI) 可使急性呼吸窘迫综合征 (ARDS) 病例更加复杂，两者结合可显著恶化预后。我们的目标是利用机器学习 (ML) 技术构建能够快速识别 ARDS 患者发生 AKI 风险的模型。
方法：我们从 Medical Information Mart for Intensive Care III (MIMIC-III) 和 MIMIC-IV 数据库中获取 ARDS 患者的数据。在 MIMIC-III 数据集内，我们开发了 11 个 ML 预测模型。通过评估各种指标，我们使用 Shapley 加性解释 (SHAP) 可视化了其特征的重要性。然后，我们使用较少的变量创建了一个更简洁的模型，并使用超参数优化 (HPO) 对其进行了优化。我们使用 MIMIC-IV 数据集对模型进行了验证。
结果：MIMIC-III 数据集共纳入 928 例无 AKI 的 ARDS 患者，其中 179 例（19.3%）在入住重症监护病房（ICU）后发生 AKI。MIMIC-IV 数据集共纳入 653 例 ARDS 患者，其中 237 例（36.3%）发生 AKI。共使用 43 个特征构建模型。所有模型中，eXtreme gradient boosting (XGBoost) 表现最佳。我们使用前 10 个特征构建了一个 AUC 为 0.850 的紧凑模型，经 HPO 优化后 AUC 提高至 0.865。在额外的验证集中，XGBoost_HPO 的 AUC 为 0.854。XGBoost_HPO 模型在测试集上的准确率、敏感度、特异度、阳性预测值 (PPV)、阴性预测值 (NPV) 和 F1 评分分别为 0.865、0.813、0.877、0.578、0.957 和 0.675。在额外的验证集中，它们分别为 0.724、0.789、0.688、0.590、0.851 和 0.675。
结论：ML 算法，尤其是 XGBoost，可可靠预测 ARDS 患者 AKI 的发生。简洁模型保持了出色的预测能力，而基于网络的计算器提高了临床便利性。这为识别 ARDS 中的 AKI 提供了有价值的指导，从而改善了患者的预后。
Zhonghua Wei Zhong Bing Ji Jiu Yi Xue. 2022-11
Heart Surg Forum. 2023-10-25
J Inflamm Res. 2025-5-19
Pediatr Rev. 2023-5-1
Biomed Pharmacother. 2023-3
Sci Data. 2023-1-3
Molecules. 2022-10-28
Zotero 插件
