Department of Emergency Medicine, Seoul National University Bundang Hospital, 166 Gumi-ro, Bundang-gu, Gyeonggi-do, Seongnam-si 463-707, Republic of Korea.
Department of Emergency Medicine, CHA Bundang Medical Center, CHA University, 59, Yatap-ro, Bundang-gu, Gyeonggi-do, Seongnam-si 463-712, Republic of Korea.
J Crit Care. 2020 Feb;55:163-170. doi: 10.1016/j.jcrc.2019.09.024. Epub 2019 Oct 22.
We hypothesized utilizing machine learning (ML) algorithms for screening septic shock in ED would provide better accuracy than qSOFA or MEWS.
The study population was adult (≥20 years) patients visiting ED for suspected infection. Target event was septic shock within 24 h after arrival. Demographics, vital signs, level of consciousness, chief complaints (CC) and initial blood test results were used as predictors. CC were embedded into 16-dimensional vector space using singular value decomposition. Six base learners including support vector machine, gradient-boosting machine, random forest, multivariate adaptive regression splines and least absolute shrinkage and selection operator and ridge regression and their ensembles were tested. We also trained and tested MLP networks with various setting.
A total of 49,560 patients were included and 4817 (9.7%) had septic shock within 24 h. All ML classifiers significantly outperformed qSOFA score, MEWS and their age-sex adjusted versions with their AUROC ranging from 0.883 to 0.929. The ensembles of the base classifiers showed the best performance and addition of CC embedding was associated with statistically significant increases in performance.
ML classifiers significantly outperforms clinical scores in screening septic shock at ED triage.
我们假设利用机器学习（ML）算法对 ED 中的脓毒症休克进行筛选，其准确性将优于 qSOFA 或 MEWS。
研究人群为因疑似感染而就诊 ED 的成年（≥20 岁）患者。目标事件为到达后 24 小时内发生的脓毒症休克。使用人口统计学、生命体征、意识水平、主要症状（CC）和初始血液检查结果作为预测因素。CC 使用奇异值分解嵌入到 16 维向量空间中。测试了包括支持向量机、梯度提升机、随机森林、多变量自适应回归样条和最小绝对值收缩和选择算子以及岭回归及其集成在内的 6 个基本学习器。我们还使用不同的设置训练和测试了 MLP 网络。
共纳入 49560 例患者，其中 4817 例（9.7%）在 24 小时内发生脓毒症休克。所有 ML 分类器均显著优于 qSOFA 评分、MEWS 及其年龄性别调整版本，其 AUC 范围为 0.883 至 0.929。基本分类器的集成表现最佳，并且 CC 嵌入的添加与性能的统计学显著提高相关。
ML 分类器在 ED 分诊中筛查脓毒症休克的表现明显优于临床评分。
