机器学习模型预测急诊分诊临床结局
该研究利用2007至2015年NHAMCS急诊数据,纳入年龄≥18岁的成人就诊记录,以分诊时常规可获数据(人口学、生命体征、主诉、合并症)为预测因子,在随机抽取的70%训练集中构建Lasso回归、随机森林、梯度提升决策树和深度神经网络四种机器学习模型,并以五级ESI数据构建逻辑回归作为参考模型;在剩余30%测试集中评估。135,470次就诊中,2.1%发生危重症结局、16.2%发生住院结局。危重症预测中深度神经网络AUC 0.86(95%CI 0.85-0.87)优于参考模型0.74(95%CI 0.72-0.75);住院预测中深度神经网络AUC 0.82(95%CI 0.82-0.83)优于参考模型0.69(95%CI 0.68-0.69);所有机器学习模型在决策曲线分析中各临床阈值范围内净获益均更高。作者结论认为机器学习模型预测表现优于传统方法,可能增强分诊决策。
要点
- 危重症结局预测:深度神经网络AUC 0.86(95%CI 0.85-0.87),参考模型AUC 0.74(95%CI 0.72-0.75)。
- 住院结局预测:深度神经网络AUC 0.82(95%CI 0.82-0.83),参考模型AUC 0.69(95%CI 0.68-0.69)。
- 危重症预测中,机器学习模型在ESI 3至5级(紧急至非紧急)漏诊(under-triage)患者更少。
- 住院预测中,机器学习模型在ESI 1至3级(立即至紧急)过度分诊(over-triage)更少。
- 决策曲线分析显示所有机器学习模型在各临床阈值范围内净获益均高于参考模型。
适用人群与场景
- 2007年至2015年NHAMCS急诊数据中年龄≥18岁的成人就诊记录,共135,470次符合条件就诊。
诊断标准与关键阈值
- 以分诊时常规可获数据(人口学、分诊生命体征、主诉、合并症)为预测因子。
- 危重症结局定义为入住重症监护室或院内死亡。
- 住院结局定义为直接住院或转院。
- 参考模型使用五级ESI数据构建逻辑回归。
研究设计与关键结果
- 研究类型:利用NHAMCS数据的机器学习模型开发与验证研究;随机抽取70%为训练集,剩余30%为测试集。
- 135,470次就诊中,2.1%发生危重症结局,16.2%发生住院结局。
- 危重症预测:深度神经网络AUC 0.86(95%CI 0.85-0.87) vs 参考模型AUC 0.74(95%CI 0.72-0.75)。
- 住院预测:深度神经网络AUC 0.82(95%CI 0.82-0.83) vs 参考模型AUC 0.69(95%CI 0.68-0.69)。
- 危重症预测中,机器学习模型在ESI 3至5级(紧急至非紧急)漏诊患者更少。
- 住院预测中,机器学习模型在ESI 1至3级(立即至紧急)过度分诊患者更少。
- 决策曲线分析中,所有机器学习模型在各临床阈值范围内净获益均更高。
证据类型、等级与局限
- 证据类型:基于NHAMCS数据的机器学习模型开发与验证研究。
- 局限:原文未报告局限性。
对床旁实践的改变
- 作者认为,应用现代机器学习模型可能增强临床医生的分诊决策,从而实现更好的临床照护和资源优化。
提炼说明
- 本条依据论文摘要整理;摘要未涉及「处置流程与药物剂量」「红旗征与不能漏的情况」「指南或共识出处」,故未列出这些栏目;全文请点「阅读原文」。
英文原文(摘要)· 点击展开
BACKGROUND: Development of emergency department (ED) triage systems that accurately differentiate and prioritize critically ill from stable patients remains challenging. We used machine learning models to predict clinical outcomes, and then compared their performance with that of a conventional approach-the Emergency Severity Index (ESI).
METHODS: Using National Hospital and Ambulatory Medical Care Survey (NHAMCS) ED data, from 2007 through 2015, we identified all adult patients (aged ≥ 18 years). In the randomly sampled training set (70%), using routinely available triage data as predictors (e.g., demographics, triage vital signs, chief complaints, comorbidities), we developed four machine learning models: Lasso regression, random forest, gradient boosted decision tree, and deep neural network. As the reference model, we constructed a logistic regression model using the five-level ESI data. The clinical outcomes were critical care (admission to intensive care unit or in-hospital death) and hospitalization (direct hospital admission or transfer). In the test set (the remaining 30%), we measured the predictive performance, including area under the receiver-operating-characteristics curve (AUC) and net benefit (decision curves) for each model.
RESULTS: Of 135,470 eligible ED visits, 2.1% had critical care outcome and 16.2% had hospitalization outcome. In the critical care outcome prediction, all four machine learning models outperformed the reference model (e.g., AUC, 0.86 [95%CI 0.85-0.87] in the deep neural network vs 0.74 [95%CI 0.72-0.75] in the reference model), with less under-triaged patients in ESI triage levels 3 to 5 (urgent to non-urgent). Likewise, in the hospitalization outcome prediction, all machine learning models outperformed the reference model (e.g., AUC, 0.82 [95%CI 0.82-0.83] in the deep neural network vs 0.69 [95%CI 0.68-0.69] in the reference model) with less over-triages in ESI triage levels 1 to 3 (immediate to urgent). In the decision curve analysis, all machine learning models consistently achieved a greater net benefit-a larger number of appropriate triages considering a trade-off with over-triages-across the range of clinical thresholds.
CONCLUSIONS: Compared to the conventional approach, the machine learning models demonstrated a superior performance to predict critical care and hospitalization outcomes. The application of modern machine learning models may enhance clinicians' triage decision making, thereby achieving better clinical care and optimal resource utilization.
参考来源
- Raita Y, Goto T, Faridi MK, et al. Emergency department triage prediction of clinical outcomes using machine learning models. Crit Care. 2019. PMID: 30795786. doi:10.1186/s13054-019-2351-7
供执业医师抢救与诊疗参考,不替代临床判断;剂量请结合体重、肝肾功能和本院药品规格复核,以指南原文为准。