文献速递呼吸衰竭与机械通气重症综合管理与镇静镇痛

机器学习模型预测社区获得性肺炎患者死亡:开发与验证研究

Machine-Learning Model for Mortality Prediction in Patients With Community-Acquired Pneumonia: Development and Validation Study.
PubMed 文献 · Chest · 2022-07-16
一句话结论在两家西班牙大学医院中,SeF-ML预测CAP患者30天死亡的AUC为0.826,显著高于CURB-65和qSOFA,但与PSI和SOFA无显著差异。

本研究为在两家西班牙大学医院开展的回顾性推导-验证研究,旨在评估一种为脓毒症设计并适配CAP的因果概率网络模型(SeF-ML)预测30天死亡的能力,并与PSI、SOFA、qSOFA和CURB-65比较。推导队列纳入4,531例患者,验证队列1,034例。验证队列中SeF-ML的AUC为0.826,与推导队列的0.801一致(P = .51);SeF-ML的AUC显著高于CURB-65(0.764;P = .03)和qSOFA(0.729;P = .005),但与PSI(0.830;P = .92)和SOFA(0.771;P = .14)无显著差异。作者认为SeF-ML在利用结构化健康数据改善CAP死亡预测方面显示出潜力,并建议开展更多外部验证研究以支持其泛化性。

要点

  • 验证队列中SeF-ML预测30天死亡的AUC为0.826,与推导队列的0.801一致(P = .51)。
  • SeF-ML的AUC显著高于CURB-65(0.764;P = .03)和qSOFA(0.729;P = .005)。
  • SeF-ML与PSI(0.830;P = .92)和SOFA(0.771;P = .14)的AUC无显著差异。
  • SeF模型为专有软件,使用结构化健康数据预测CAP患者死亡。
  • 作者认为SeF-ML在改善CAP死亡预测方面显示出潜力,建议开展更多外部验证研究以支持泛化性。

适用人群与场景

  • 社区获得性肺炎(CAP)患者,来自两家西班牙大学医院。
  • 推导队列4,531例患者,验证队列1,034例患者。

研究设计与关键结果

  • 研究类型:回顾性推导-验证研究,在两家西班牙大学医院开展。
  • 干预/比较:评估为脓毒症设计并适配CAP的因果概率网络模型(SeF-ML)预测30天死亡的能力,并与PSI、SOFA、qSOFA和CURB-65比较。
  • 推导队列(4,531例)30天死亡预测AUC:SeF-ML 0.801,CURB-65 0.759,SOFA 0.671,PSI 0.799,qSOFA 0.642。
  • 验证队列(1,034例)30天死亡预测AUC:SeF-ML 0.826,与推导队列的0.801一致(P = .51)。
  • 验证队列中SeF-ML的AUC显著高于CURB-65(0.764;P = .03)和qSOFA(0.729;P = .005)。
  • 验证队列中SeF-ML的AUC与PSI(0.830;P = .92)和SOFA(0.771;P = .14)无显著差异。

证据类型、等级与局限

  • 证据类型:回顾性推导-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SeF-ML在利用结构化健康数据改善CAP患者死亡预测方面显示出潜力。
  • 作者建议开展更多外部验证研究以支持SeF-ML的泛化性。

提炼说明

  • 本条依据论文摘要整理;摘要未涉及「诊断标准与关键阈值」「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: Artificial intelligence tools and techniques such as machine learning (ML) are increasingly seen as a suitable manner in which to increase the prediction capacity of currently available clinical tools, including prognostic scores. However, studies evaluating the efficacy of ML methods in enhancing the predictive capacity of existing scores for community-acquired pneumonia (CAP) are limited. We aimed to apply and validate a causal probabilistic network (CPN) model to predict mortality in patients with CAP.

RESEARCH QUESTION: Is a CPN model able to predict mortality in patients with CAP better than the commonly used severity scores?

STUDY DESIGN AND METHODS: This was a derivation-validation retrospective study conducted in two Spanish university hospitals. The ability of a CPN designed to predict mortality in sepsis (SepsisFinder [SeF]), and adapted for CAP (SeF-ML), to predict 30-day mortality was assessed and compared with other scoring systems (Pneumonia Severity Index [PSI], Sequential Organ Failure Assessment [SOFA], quick Sequential Organ Failure Assessment [qSOFA], and CURB-65 criteria [confusion, urea, respiratory rate, BP, age ≥ 65 years]). The SeF models are proprietary software. Differences between receiver operating characteristic curves were assessed by the DeLong method for correlated receiver operating characteristic curves.

RESULTS: The derivation cohort comprised 4,531 patients, and the validation cohort consisted of 1,034 patients. In the derivation cohort, the areas under the curve (AUCs) of SeF-ML, CURB-65, SOFA, PSI, and qSOFA were 0.801, 0.759, 0.671, 0.799, and 0.642, respectively, for 30-day mortality prediction. In the validation study, the AUC of SeF-ML was 0.826, concordant with the AUC (0.801) in the derivation data (P = .51). The AUC of SeF-ML was significantly higher than those of CURB-65 (0.764; P = .03) and qSOFA (0.729, P = .005). However, it did not differ significantly from those of PSI (0.830; P = .92) and SOFA (0.771; P = .14).

INTERPRETATION: SeF-ML shows potential for improving mortality prediction among patients with CAP, using structured health data. Additional external validation studies should be conducted to support generalizability.

来源:Chest,查看原文。版权归原作者所有。

参考来源

  1. Cilloniz C, Ward L, Mogensen ML, et al. Machine-Learning Model for Mortality Prediction in Patients With Community-Acquired Pneumonia: Development and Validation Study. Chest. 2022. PMID: 35850287. doi:10.1016/j.chest.2022.07.005

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。