Early identification of emergency department (ED) patients who need hospitalization is essential for quality of care and patient safety. We aimed to compare machine learning (ML) models predicting the hospitalization of ED patients and conventional regression techniques at three points in time after ED registration. We analyzed consecutive ED patients of three hospitals using the Netherlands Emergency Department Evaluation Database (NEED). We developed prediction models for hospitalization using an increasing number of data available at triage, ∼30 min (including vital signs) and ∼2 h (including laboratory tests) after ED registration, using ML (random forest, gradient boosted decision trees, deep neural networks) and multivariable logistic regression analysis (including spline transformations for continuous predictors). Demographics, urgency, presenting complaints, disease severity and proxies for comorbidity, and complexity were used as covariates. We compared the performance using the area under the ROC curve in independent validation sets from each hospital. We included 172,104 ED patients of whom 66,782 (39 %) were hospitalized. The AUC of the multivariable logistic regression model was 0.82 (0.78−0.86) at triage, 0.84 (0.81−0.86) at ∼30 min and 0.83 (0.75−0.92) after ∼2 h. The best performing ML model over time was the gradient boosted decision trees model with an AUC of 0.84 (0.77−0.88) at triage, 0.86 (0.82−0.89) at ∼30 min and 0.86 (0.74−0.93) after ∼2 h. Our study showed that machine learning models had an excellent but similar predictive performance as the logistic regression model for predicting hospital admission. In comparison to the 30-min model, the 2-h model did not show a performance improvement. After further validation, these prediction models could support management decisions by real-time feedback to medical personal.

中文翻译：

---

机器学习用于开发急诊科患者入院预测模型：炒作还是希望？

及早识别需要住院的急诊科 (ED) 患者对于护理质量和患者安全至关重要。我们的目的是在 ED 登记后的三个时间点比较预测 ED 患者住院的机器学习 (ML) 模型和传统回归技术。我们使用荷兰急诊科评估数据库（NEED）对三家医院的连续急诊患者进行了分析。我们使用 ML（随机森林、梯度增强决策树、深度神经网络）开发住院预测模型，使用越来越多的分诊、急诊登记后 30 分钟（包括生命体征）和 2 小时（包括实验室检查）可用的数据。网络）和多变量逻辑回归分析（包括连续预测变量的样条变换）。人口统计、紧迫性、主诉、疾病严重程度和合并症代理以及复杂性被用作协变量。我们使用每家医院的独立验证集中的 ROC 曲线下面积来比较性能。我们纳入了 172,104 名 ED 患者，其中 66,782 名 (39%) 住院。多变量逻辑回归模型的 AUC 在分诊时为 0.82 (0.78−0.86)，在 ~30 分钟时为 0.84 (0.81−0.86)，在 ~2 小时后为 0.83 (0.75−0.92)。随着时间的推移，表现最好的 ML 模型是梯度增强决策树模型，分类时的 AUC 为 0.84 (0.77−0.88)，~30 分钟时的 AUC 为 0.86 (0.82−0.89)，~2 小时后为 0.86 (0.74−0.93)。我们的研究表明，机器学习模型与预测入院的逻辑回归模型具有出色但相似的预测性能。与 30 分钟模型相比，2 小时模型没有表现出性能改进。经过进一步验证，这些预测模型可以通过实时反馈给医务人员来支持管理决策。