Department of Medical Informatics, Tohoku University Graduate School of Medicine, Miyagi, Japan.
PLoS One. 2023 Sep 21;18(9):e0291711. doi: 10.1371/journal.pone.0291711. eCollection 2023.
The aim of this study was to develop early prediction models for respiratory failure risk in patients with severe pneumonia using four ensemble learning algorithms: LightGBM, XGBoost, CatBoost, and random forest, and to compare the predictive performance of each model. In this study, we used the eICU Collaborative Research Database (eICU-CRD) for sample extraction, built a respiratory failure risk prediction model for patients with severe pneumonia based on four ensemble learning algorithms, and developed compact models corresponding to the four complete models to improve clinical practicality. The average area under receiver operating curve (AUROC) of the models on the test sets after ten random divisions of the dataset and the average accuracy at the best threshold were used as the evaluation metrics of the model performance. Finally, feature importance and Shapley additive explanation values were introduced to improve the interpretability of the model. A total of 1676 patients with pneumonia were analyzed in this study, of whom 297 developed respiratory failure one hour after admission to the intensive care unit (ICU). Both complete and compact CatBoost models had the highest average AUROC (0.858 and 0.857, respectively). The average accuracies at the best threshold were 75.19% and 77.33%, respectively. According to the feature importance bars and summary plot of the predictor variables, activetx (indicates whether the patient received active treatment), standard deviation of prothrombin time-international normalized ratio, Glasgow Coma Scale verbal score, age, and minimum oxygen saturation and respiratory rate were important. Compared with other ensemble learning models, the complete and compact CatBoost models have significantly higher average area under the curve values on the 10 randomly divided test sets. Additionally, the standard deviation (SD) of the compact CatBoost model is relatively small (SD:0.050), indicating that the performance of the compact CatBoost model is stable among these four ensemble learning models. The machine learning predictive models built in this study will help in early prediction and intervention of respiratory failure risk in patients with pneumonia in the ICU.
本研究旨在使用四种集成学习算法（LightGBM、XGBoost、CatBoost 和随机森林）开发重症肺炎患者呼吸衰竭风险的早期预测模型，并比较各模型的预测性能。本研究使用 eICU 协作研究数据库（eICU-CRD）进行样本提取，基于四种集成学习算法为重症肺炎患者构建呼吸衰竭风险预测模型，并开发与四个完整模型对应的精简模型，以提高临床实用性。在数据集十次随机划分后的测试集上，模型的平均接收者操作特征曲线下面积（AUROC）和最佳阈值处的平均准确率作为模型性能的评价指标。最后，引入特征重要性和 Shapley 加性解释值以提高模型的可解释性。本研究共分析了 1676 例肺炎患者，其中 297 例在入住重症监护病房（ICU）后 1 小时内发生呼吸衰竭。完整和精简的 CatBoost 模型的平均 AUROC 最高（分别为 0.858 和 0.857）。最佳阈值处的平均准确率分别为 75.19%和 77.33%。根据预测变量的特征重要性栏和汇总图，activetx（表示患者是否接受积极治疗）、凝血酶原时间国际标准化比值的标准差、格拉斯哥昏迷量表言语评分、年龄、最低氧饱和度和呼吸频率是重要的。与其他集成学习模型相比，完整和精简的 CatBoost 模型在 10 个随机划分的测试集中的平均 AUC 值明显更高。此外，精简的 CatBoost 模型的标准差（SD）较小（SD：0.050），这表明在这四种集成学习模型中，精简的 CatBoost 模型的性能稳定。本研究构建的机器学习预测模型将有助于对 ICU 中肺炎患者的呼吸衰竭风险进行早期预测和干预。
