Helleberg Johan, Sundelin Anna, Mårtensson Johan, Rooyackers Olav, Thobaben Ragnar
Perioperative Medicine and Intensive Care, Karolinska University Hospital, Stockholm, Sweden.
Section of Anaesthesiology and Intensive Care Medicine, Department of Clinical Sciences, Intervention and Technology (CLINTEC), Karolinska Institutet, Stockholm, Sweden.
BMC Med Inform Decis Mak. 2025 Jul 24;25(1):275. doi: 10.1186/s12911-025-03115-3.
In the Intensive Care Unit (ICU), data stored in patient data management systems (PDMS) is commonly used in clinical practice and research. Parameters from point-of-care arterial blood gas (BG) analysis are used in the diagnosis and definition of syndromes such as sepsis and ARDS, but manual entry of the blood source (arterial or venous) into the PDMS introduces the risk of mislabeling venous samples as arterial. Our study aimed to employ supervised machine learning to accurately identify blood gas samples as arterial or venous using PDMS data.
A retrospective, single-center observational cohort study including all blood gases during 2018 from a Swedish, pediatric and adult general ICU. Chemical parameters from BG analysis and clinical parameters such as mean arterial pressure (MAP) and saturation (SpO2) were utilized as features. A specialist physician in Intensive Care manually determined the true class of each sample through comprehensive retrospective chart review. The samples were split into training, testing and holdout sets. Training was performed using cross-validation in the training set, with forward stepwise feature selection and Bayesian hyperparameter optimization, and accuracy was assessed using area under the precision recall curve (AUCPR) in the test set. The best model was compared to a multivariate logistic regression model (LR) in the holdout set.
Among 33,800 samples (30,753 arterial, 3,047 non-arterial) from 691 ICU admissions, 150 (0.44%) were erroneously marked. The best performing algorithm was extreme gradient boosting (XGboost) using 9 features, with an AUCPR of 0.9974 (95% CI 0.9961-0.9984), significantly better than the LR model (AUCPR = 0.9791, 95% CI 0.9651-0.9904).
Supervised machine learning demonstrates efficacy in determining blood gas sample type from ICU patients. This approach shows promise for improving the accuracy of research and clinical applications relying on blood gas data.
在重症监护病房（ICU）中，患者数据管理系统（PDMS）中存储的数据常用于临床实践和研究。即时护理动脉血气（BG）分析的参数用于脓毒症和急性呼吸窘迫综合征（ARDS）等综合征的诊断和定义，但手动将血样来源（动脉或静脉）输入PDMS会带来将静脉样本误标记为动脉样本的风险。我们的研究旨在利用监督式机器学习，使用PDMS数据准确识别血气样本是动脉血还是静脉血。
一项回顾性、单中心观察性队列研究，纳入了2018年瑞典儿科和成人综合ICU的所有血气样本。BG分析的化学参数以及平均动脉压（MAP）和血氧饱和度（SpO2）等临床参数被用作特征。一名重症监护专科医生通过全面的回顾性病历审查手动确定每个样本的真实类别。样本被分为训练集、测试集和留存集。在训练集中使用交叉验证进行训练，采用前向逐步特征选择和贝叶斯超参数优化，并在测试集中使用精确召回率曲线下面积（AUCPR）评估准确性。在留存集中将最佳模型与多元逻辑回归模型（LR）进行比较。
在691例ICU入院患者的33800个样本（30753个动脉血样本，3047个非动脉血样本）中，有150个（0.44%）被错误标记。表现最佳的算法是使用9个特征的极端梯度提升（XGboost），AUCPR为0.9974（95%CI 0.9961 - 0.9984），显著优于LR模型（AUCPR = 0.9791，95%CI 0.9651 - 0.9904）。
监督式机器学习在确定ICU患者血气样本类型方面显示出有效性。这种方法有望提高依赖血气数据的研究和临床应用的准确性。
