Machine Learning Prediction Models for Mechanically Ventilated Patients: Analyses of the MIMIC-III Database

Frontiers in Medicine

article

Yibing Zhu¹  Hua Zheng³  Yan Chen³  Qianqian Guo⁴  Lin Li⁵  Bin Du³  Xiuming Xi⁶  Wei Li¹  Huibin Huang⁷  Yang Li⁴  Qian Yu⁴  Jin Zhang⁸  Guowei Wang⁵  Renqi Yao⁹  Chao Ren^1,10  Ge Chen¹  Xin Jin^1,11  Junyang Guo^1,12  Shi Liu^1,13 

[1] Medical Research and Biometrics Center, National Center for Cardiovascular Diseases, Fuwai Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College;Department of Emergency, Guang'anmen Hospital, China Academy of Chinese Medical Sciences;Peking Union Medical College Hospital, Peking Union Medical College and Chinese Academy of Medical Sciences;Department of Anesthesiology, Peking University Shougang Hospital;School of Computer Science and Technology, Wuhan University of Technology;Department of Critical Care Medicine, Fuxing Hospital, Capital Medical University;Department of Critical Care Medicine, Beijing Tsinghua Changgung Hospital, School of Clinical Medicine, Tsinghua University;School of Economics and Management, Beijing Institute of Technology;Department of Burn Surgery, The First Affiliated Hospital of Naval Medical University;Translational Medicine Research Center, Fourth Medical Center and Medical Innovation Research Division of the Chinese People's Liberation Army (PLA) General Hospital;Yidu Cloud Technology Inc.;Ltd.;School of Information Science and Engineering, Hebei North University

关键词: prediction model;    machine learning;    mechanical ventilation;    intensive care unit;    death;   

DOI  :  10.3389/fmed.2021.662340

学科分类：社会科学、人文和艺术（综合）

来源: Frontiers

【 摘 要 】

Background: Mechanically ventilated patients in the intensive care unit (ICU) have high mortality rates. There are multiple prediction scores, such as the Simplified Acute Physiology Score II (SAPS II), Oxford Acute Severity of Illness Score (OASIS), and Sequential Organ Failure Assessment (SOFA), widely used in the general ICU population. We aimed to establish prediction scores on mechanically ventilated patients with the combination of these disease severity scores and other features available on the first day of admission. Methods: A retrospective administrative database study from the Medical Information Mart for Intensive Care (MIMIC-III) database was conducted. The exposures of interest consisted of the demographics, pre-ICU comorbidity, ICU diagnosis, disease severity scores, vital signs, and laboratory test results on the first day of ICU admission. Hospital mortality was used as the outcome. We used the machine learning methods of k -nearest neighbors (KNN), logistic regression, bagging, decision tree, random forest, Extreme Gradient Boosting (XGBoost), and neural network for model establishment. A sample of 70% of the cohort was used for the training set; the remaining 30% was applied for testing. Areas under the receiver operating characteristic curves (AUCs) and calibration plots would be constructed for the evaluation and comparison of the models' performance. The significance of the risk factors was identified through models and the top factors were reported. Results: A total of 28,530 subjects were enrolled through the screening of the MIMIC-III database. After data preprocessing, 25,659 adult patients with 66 predictors were included in the model analyses. With the training set, the models of KNN, logistic regression, decision tree, random forest, neural network, bagging, and XGBoost were established and the testing set obtained AUCs of 0.806, 0.818, 0.743, 0.819, 0.780, 0.803, and 0.821, respectively. The calibration curves of all the models, except for the neural network, performed well. The XGBoost model performed best among the seven models. The top five predictors were age, respiratory dysfunction, SAPS II score, maximum hemoglobin, and minimum lactate. Conclusion: The current study indicates that models with the risk of factors on the first day could be successfully established for predicting mortality in ventilated patients. The XGBoost model performs best among the seven machine learning models.

【 授权许可】

CC BY