Yuan Jia, Xiong Jiong, Yang Jinfeng, Dong Qi, Wang Yin, Cheng Yumei, Chen Xianjun, Liu Ying, Xiao Chuan, Tao Junlin, Lizhang Shuangzi, Liujiao Yangzi, Chen Qimin, Shen Feng
Department of Intensive Care Unit, The Affiliated Hospital of Guizhou Medical University, No. 28, Guiyi Road, Yunyan District, Guiyang, Guizhou 550001, China.
Guizhou Medical University, Guiyang, Guizhou, 550004, China.
Comput Methods Programs Biomed. 2025 Mar;260:108589. doi: 10.1016/j.cmpb.2025.108589. Epub 2025 Jan 6.
The growing population of elderly neurocritically ill patients highlights the need for effective prognosis prediction tools. This study aims to develop and validate machine learning (ML) models for predicting 28-day mortality in intensive care units (ICUs).
Data were extracted from the Medical Information Mart for Intensive Care IV(MIMIC-IV) database, focusing on elderly neurocritical ill patients with ICU stays ≥ 24 h. The cohort was split into 70 % for training and 30 % for internal validation. We analyzed 58 variables, including demographics, vital signs, medications, lab results, comorbidities, and medical scores, using Lasso regression to identify predictors of 28-day mortality. Seven ML algorithms were evaluated, and the best model was validated with data from Guizhou Medical University Affiliated Hospital. A log-rank test was used to assess survival differences in Kaplan-Meier curves. Shapley Additive Explanations (SHAP) were used to interpret the best model, while subgroup analysis identified variations in model performance across different populations.
The study included 1,773 elderly neurocritically ill patients, with a 28-day mortality rate of 28.6 %. The Light Gradient Boosting Machine (LightGBM) outperformed other models, achieving an area under the curve (AUC) of 0.896 in internal validation and 0.812 in external validation. Kaplan-Meier analysis showed that higher LightGBM prediction scores correlated with lower survival probabilities. Key predictors identified through SHAP analysis included partial pressure of arterial carbon dioxide (PaCO), Acute physiology and chronic health evaluation II (APACHE II), white blood cell count, age, and lactate. The LightGBM model demonstrated consistent performance across various subgroups.
The LightGBM model effectively predicts 28-day mortality risk in elderly neurocritically ill patients, aiding clinicians in management and resource allocation. Its reliable performance across diverse subgroups underscores its clinical utility.
老年神经危重症患者数量不断增加，凸显了对有效预后预测工具的需求。本研究旨在开发和验证用于预测重症监护病房（ICU）患者28天死亡率的机器学习（ML）模型。
从重症监护医学信息集市IV（MIMIC-IV）数据库中提取数据，重点关注在ICU住院≥24小时的老年神经危重症患者。该队列分为70%用于训练，30%用于内部验证。我们分析了58个变量，包括人口统计学、生命体征、药物、实验室检查结果、合并症和医学评分，使用套索回归来识别28天死亡率的预测因素。评估了七种ML算法，并使用贵州医科大学附属医院的数据对最佳模型进行了验证。使用对数秩检验评估Kaplan-Meier曲线中的生存差异。使用Shapley加法解释（SHAP）来解释最佳模型，而亚组分析确定了不同人群中模型性能的差异。
该研究纳入了1773例老年神经危重症患者，28天死亡率为28.6%。轻梯度提升机（LightGBM）的表现优于其他模型，内部验证的曲线下面积（AUC）为0.896，外部验证的AUC为0.812。Kaplan-Meier分析表明，较高的LightGBM预测分数与较低的生存概率相关。通过SHAP分析确定的关键预测因素包括动脉血二氧化碳分压（PaCO）、急性生理与慢性健康状况评估II（APACHE II）、白细胞计数、年龄和乳酸。LightGBM模型在各个亚组中表现一致。
LightGBM模型有效地预测了老年神经危重症患者的28天死亡风险，有助于临床医生进行管理和资源分配。其在不同亚组中的可靠表现突出了其临床实用性。
