文献速递心脑血管急症重症综合管理与镇静镇痛

使用机器学习的颅内压预测算法(I-CARE):训练与验证研究

IntraCranial pressure prediction AlgoRithm using machinE learning (I-CARE): Training and Validation Study.
PubMed 文献 · Crit Care Explor · 2023-12-28
一句话结论基于电子ICU数据训练的机器学习模型可提前30分钟预测颅内压,eICU测试集均方根误差4.51 mmHg,MIMIC-III外部验证集3.56 mmHg。

本研究利用美国208家医院335个ICU的eICU协作研究数据库,训练集成机器学习模型预测成人患者30分钟后的颅内压。模型输入包括年龄、性别、实验室指标、药物与输注、出入量、格拉斯哥昏迷量表成分及时间序列生命体征,每段95分钟时间块用前12个5分钟间隔观察值预测最后观察值后30分钟的5分钟中位颅内压。931例颅内压监测患者共46,207个时间块用于训练,eICU测试集均方根误差4.51 mmHg,MIMIC-III匹配波形数据库外部验证均方根误差3.56 mmHg。作者认为该模型预测性能有前景,包括在外部验证数据集中。

要点

  • 模型预测30分钟后的5分钟中位颅内压,eICU测试集均方根误差4.51 mmHg,MIMIC-III外部验证集3.56 mmHg。
  • 训练数据来自eICU协作研究数据库931例颅内压监测成人患者,共46,207个95分钟时间块。
  • 驱动颅内压预测最重要的变量包括既往颅内压史、体温、体重、血清肌酐、年龄、格拉斯哥昏迷量表及血流动力学参数。
  • 预测因子涵盖年龄、性别、实验室指标、药物与输注、出入量、格拉斯哥昏迷量表成分及心率、颅内压、平均动脉压、呼吸频率、体温等时间序列生命体征。
  • 作者认为该集成机器学习模型在外部验证数据集中也显示出有前景的预测性能。

适用人群与场景

  • 来自电子ICU(eICU)协作研究数据库的成人患者,用于训练和测试模型。
  • 外部验证使用Medical Information Mart for Intensive Care-III(MIMIC-III)匹配波形数据库。

研究设计与关键结果

  • 研究类型为回顾性研究,数据来自美国208家医院335个ICU的eICU协作研究数据库。
  • 共提取931例接受颅内压监测患者的数据,划分为46,207个95分钟时间块。
  • 每个时间块用非时间变化协变量及12个5分钟间隔的时间变化协变量观察值,预测最后观察颅内压值后30分钟的5分钟中位颅内压。
  • eICU测试集均方根误差为4.51 mmHg,MIMIC-III外部验证数据集均方根误差为3.56 mmHg。
  • 最重要的预测变量为既往颅内压史、患者体温、体重、血清肌酐、年龄、格拉斯哥昏迷量表及血流动力学参数。

证据类型、等级与局限

  • 证据类型:回顾性研究,训练并验证集成机器学习模型。
  • 局限:原文未报告局限性。

对床旁实践的改变

  • 作者认为该模型在包括外部验证数据集在内的预测性能有前景,可帮助临床医生调整治疗并可能预防颅内压升高发作。

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

OBJECTIVES: Elevated intracranial pressure (ICP) is a potentially devastating complication of neurologic injury. Developing an ICP prediction algorithm to help the clinician adjust treatments and potentially prevent elevated ICP episodes.

DESIGN: Retrospective study.

SETTING: Three hundred thirty-five ICUs at 208 hospitals in the United States.

SUBJECTS: Adults patients from the electronic ICU (eICU) Collaborative Research Database was used to train an ensemble machine learning model to predict the ICP 30 minutes in the future. Predictive performance was evaluated using a left-out test dataset and externally evaluated on the Medical Information Mart for Intensive Care-III (MIMIC-III) Matched Waveform Database.

INTERVENTIONS: None.

MEASUREMENTS AND MAIN RESULTS: Predictors included age, assigned sex, laboratories, medications and infusions, input/output, Glasgow Coma Scale (GCS) components, and time-series vitals (heart rate, ICP, mean arterial pressure, respiratory rate, and temperature). Each patient ICU stay was divided into successive 95-minute timeblocks. For each timeblock, the model was trained on nontime-varying covariates as well as on 12 observations of time-varying covariates at 5-minute intervals and asked to predict the 5-minute median ICP 30 minutes after the last observed ICP value. Data from 931 patients with ICP monitoring in the eICU dataset were extracted (46,207 timeblocks). The root mean squared error was 4.51 mm Hg in the eICU test set and 3.56 mm Hg in the MIMIC-III dataset. The most important variables driving ICP prediction were previous ICP history, patients' temperature, weight, serum creatinine, age, GCS, and hemodynamic parameters.

CONCLUSIONS: IntraCranial pressure prediction AlgoRithm using machinE learning, an ensemble machine learning model, trained to predict the ICP of a patient 30 minutes in the future based on baseline characteristics and vitals data from the past hour showed promising predictive performance including in an external validation dataset.

来源:Crit Care Explor,查看原文。版权归原作者所有。

参考来源

  1. Fong N, Feng J, Hubbard A, et al. IntraCranial pressure prediction AlgoRithm using machinE learning (I-CARE): Training and Validation Study. Crit Care Explor. 2023. PMID: 38161734. doi:10.1097/CCE.0000000000001024

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。