Department of Neurology, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, No.1095 Jiefang Avenue, Wuhan 430030, China.
Department of Neurology, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, No.1095 Jiefang Avenue, Wuhan 430030, China.
J Neurol Sci. 2024 Jan 15;456:122849. doi: 10.1016/j.jns.2023.122849. Epub 2023 Dec 21.
Intracerebral hemorrhage (ICH) is a stroke subtype characterized by high mortality and complex post-event complications. Research has extensively covered the acute phase of ICH; however, ICU readmission determinants remain less explored. Utilizing the MIMIC-III and MIMIC-IV databases, this investigation develops machine learning (ML) models to anticipate ICU readmissions in ICH patients.
Retrospective data from 2242 ICH patients were evaluated using ICD-9 codes. Recursive feature elimination with cross-validation (RFECV) discerned significant predictors of ICU readmissions. Four ML models-AdaBoost, RandomForest, LightGBM, and XGBoost-underwent development and rigorous validation. SHapley Additive exPlanations (SHAP) elucidated the effect of distinct features on model outcomes.
ICU readmission rates were 9.6% for MIMIC-III and 10.6% for MIMIC-IV. The LightGBM model, with an AUC of 0.736 (95% CI: 0.668-0.801), surpassed other models in validation datasets. SHAP analysis revealed hydrocephalus, sex, neutrophils, Glasgow Coma Scale (GCS), specific oxygen saturation (SpO2) levels, and creatinine as significant predictors of readmission.
The LightGBM model demonstrates considerable potential in predicting ICU readmissions for ICH patients, highlighting the importance of certain clinical predictors. This research contributes to optimizing patient care and ICU resource management. Further prospective studies are warranted to corroborate and enhance these predictive insights for clinical utilization.
脑出血（ICH）是一种具有高死亡率和复杂发病后并发症的中风类型。研究已经广泛涵盖了 ICH 的急性期；然而，ICU 再入院的决定因素仍较少被探索。本研究利用 MIMIC-III 和 MIMIC-IV 数据库，开发机器学习（ML）模型来预测 ICH 患者的 ICU 再入院。
使用 ICD-9 代码评估了 2242 例 ICH 患者的回顾性数据。递归特征消除与交叉验证（RFECV）辨别出 ICU 再入院的显著预测因素。开发并严格验证了四种 ML 模型——AdaBoost、RandomForest、LightGBM 和 XGBoost。Shapley Additive exPlanations（SHAP）阐明了不同特征对模型结果的影响。
MIMIC-III 的 ICU 再入院率为 9.6%，MIMIC-IV 的再入院率为 10.6%。LightGBM 模型在验证数据集的 AUC 为 0.736（95%CI：0.668-0.801），优于其他模型。SHAP 分析表明，脑积水、性别、中性粒细胞、格拉斯哥昏迷量表（GCS）、特定血氧饱和度（SpO2）水平和肌酐是再入院的重要预测因素。
LightGBM 模型在预测 ICH 患者的 ICU 再入院方面具有相当大的潜力，突出了某些临床预测因素的重要性。本研究有助于优化患者护理和 ICU 资源管理。需要进一步的前瞻性研究来证实和增强这些预测见解，以用于临床应用。
