开发并验证一种机器学习集成模型以预测高危手术患者队列中的术后谵妄：一项二次队列分析。
Developing and validating a machine learning ensemble model to predict postoperative delirium in a cohort of high-risk surgical patients: A secondary cohort analysis.
作者信息
Neto Paulo C S, Rodrigues Attila L, Stahlschmidt Adriene, Helal Lucas, Stefani Luciana C
机构信息
From the Programa de Pós-graduação em Medicina: Ciências Médicas, Universidade Federal do Rio Grande do Sul (PCSN), Universidade Federal do Rio Grande do Sul (ALR), Programa de Pós-graduação em Medicina: Ciências Médicas, Universidade Federal do Rio Grande do Sul (AS), Hospital de Clínicas de Porto Alegre and Universidade Federal do Rio Grande do Sul (LH), Programa de Pós-graduação em Medicina: Ciências Médicas, Professor at Surgical Department -Universidade Federal do Rio Grande do Sul and Chief of Teaching Division of Hospital de Clínicas de Porto Alegre, Porto Alegre, Brazil (LCS).
出版信息
Eur J Anaesthesiol. 2023 May 1;40(5):356-364. doi: 10.1097/EJA.0000000000001811. Epub 2023 Mar 1.
BACKGROUND
Postoperative delirium (POD) has a negative impact on prognosis, length of stay and the burden of care. Although its prediction and identification may improve postoperative care, this need is largely unmet in the Brazilian public health system.
OBJECTIVE
To develop and validate a machine-learning prediction model and estimate the incidence of delirium. We hypothesised that an ensemble machine-learning prediction model that incorporates predisposing and precipitating features could accurately predict POD.
DESIGN
A secondary analysis nested in a cohort of high-risk surgical patients.
SETTING
An 800-bed, quaternary university-affiliated teaching hospital in Southern Brazil. We included patients operated on from September 2015 to February 2020.
PATIENTS
We recruited 1453 inpatients with an all-cause postoperative 30-day mortality risk greater than 5% assessed preoperatively by the ExCare Model.
MAIN OUTCOME MEASURE
The incidence of POD classified by the Confusion Assessment Method, up to 7 days postoperatively. Predictive model performance with different feature scenarios were compared with the area under the receiver operating characteristic curve.
RESULTS
The cumulative incidence of delirium was 117, giving an absolute risk of 8.05/100 patients. We developed multiple machine-learning nested cross-validated ensemble models. We selected features through partial dependence plot analysis and theoretical framework. We treated the class imbalance with undersampling. Different feature scenarios included: 52 preoperative, 60 postoperative and only three features (age, preoperative length of stay and the number of postoperative complications). The mean areas (95% confidence interval) under the curve ranged from 0.61 (0.59 to 0.63) to 0.74 (0.73 to 0.75).
CONCLUSION
A predictive model composed of three indicative readily available features performed better than those with numerous perioperative features, pointing to its feasibility as a prognostic tool for POD. Further research is required to test the generalisability of this model.
TRIAL REGISTRATION
Institutional Review Board Registration number 04448018.8.0000.5327 (Brazilian CEP/CONEP System, available in https://plataformabrasil.saude.gov.br/ ).
背景
术后谵妄（POD）对预后、住院时间和护理负担有负面影响。尽管其预测和识别可能改善术后护理，但在巴西公共卫生系统中，这一需求在很大程度上未得到满足。
目的
开发并验证一种机器学习预测模型，并估计谵妄的发生率。我们假设，一个整合了易患因素和促发因素特征的集成机器学习预测模型能够准确预测POD。
设计
对一组高危手术患者队列进行的二次分析。
设置
巴西南部一家拥有800张床位的大学附属四级教学医院。我们纳入了2015年9月至2020年2月期间接受手术的患者。
患者
我们招募了1453名住院患者，这些患者术前经ExCare模型评估全因术后30天死亡风险大于5%。
主要观察指标
采用意识模糊评估法分类的POD发生率，术后长达7天。将不同特征情景下预测模型的性能与受试者操作特征曲线下面积进行比较。
结果
谵妄的累积发生率为117例，绝对风险为8.05/100例患者。我们开发了多个机器学习嵌套交叉验证的集成模型。我们通过部分依赖图分析和理论框架选择特征。我们采用欠采样处理类别不平衡问题。不同的特征情景包括：52个术前特征、60个术后特征以及仅三个特征（年龄、术前住院时间和术后并发症数量）。曲线下平均面积（95%置信区间）范围为0.61（0.59至0.63）至0.74（0.73至0.75）。
结论
由三个易于获得的指示性特征组成的预测模型比具有众多围手术期特征的模型表现更好，表明其作为POD预后工具的可行性。需要进一步研究来检验该模型的通用性。
试验注册
机构审查委员会注册号04448018.8.0000.5327（巴西CEP/CONEP系统，可在https://plataformabrasil.saude.gov.br/获取）。
Zotero 插件