Supervised machine learning for the early prediction of acute respiratory distress syndrome (ARDS)
This article has been Reviewed by the following groups
Listed in
- Evaluated articles (ScreenIT)
Abstract
No abstract available
Article activity feed
-
-
SciScore for 10.1101/2020.03.19.20038364: (What is this?)
Please note, not all rigor criteria are appropriate for all manuscripts.
Table 1: Rigor
Institutional Review Board Statement IRB: Data selection: Data were obtained from the Medical Information Mart for Intensive Care III (MIMIC-III) database, which consists of the inpatient ICU encounters at Beth Israel Deaconess Medical Center between 2001 and 2012.46 The MIMIC-III publication states that, “the project was approved by the Institutional Review Boards of Beth Israel Deaconess Medical Center (Boston, MA) and the Massachusetts Institute of Technology (Cambridge, MA).
Consent: Requirement for individual patient consent was waived because the project did not impact clinical care and all protected health information was deidentified.”46 To ensure consistent encoding of data, only data …SciScore for 10.1101/2020.03.19.20038364: (What is this?)
Please note, not all rigor criteria are appropriate for all manuscripts.
Table 1: Rigor
Institutional Review Board Statement IRB: Data selection: Data were obtained from the Medical Information Mart for Intensive Care III (MIMIC-III) database, which consists of the inpatient ICU encounters at Beth Israel Deaconess Medical Center between 2001 and 2012.46 The MIMIC-III publication states that, “the project was approved by the Institutional Review Boards of Beth Israel Deaconess Medical Center (Boston, MA) and the Massachusetts Institute of Technology (Cambridge, MA).
Consent: Requirement for individual patient consent was waived because the project did not impact clinical care and all protected health information was deidentified.”46 To ensure consistent encoding of data, only data collected with the MetaVision clinical information system were used.Randomization In this validation paradigm, the data was partitioned into ten random segments, or folds. Blinding not detected. Power Analysis not detected. Sex as a biological variable not detected. Table 2: Resources
Software and Algorithms Sentences Resources All predictive models described in this paper are instances of the XGBoost gradient boosted tree model,54 implemented using the Python package. Pythonsuggested: (IPython, RRID:SCR_001658)Five folds for hyperparameter tuning is the default for hyperparameter grid search due to considerations of computational constraints, as implemented in Scikit-learn. Scikit-learnsuggested: (scikit-learn, RRID:SCR_002577)Results from OddPub: We did not detect open data. We also did not detect open code. Researchers are encouraged to share open data when possible (see Nature blog).
Results from LimitationRecognizer: We detected the following sentences addressing limitations in the study:The use of this structured data to complement radiographic reports softens the limitations on timeliness which may be presented by requiring the use of information derived from chest radiographs. In these ways, the method we describe diversifies and improves upon existing approaches for the prediction of ARDS. Inability to anticipate which patients are likely to develop ARDS is a major obstacle to early intervention or prevention studies.56 Epidemiologic data suggest that the syndrome is rarely present at the time of hospital admission or initial emergency department (ED) evaluation, but develops over a period of hours to days in subsets of at-risk patients.57-61 Therefore, evaluating model performance at >24 hours preceding onset is valuable because it facilitates identification of patients who would benefit from ARDS progression and prevention interventions. Alerting systems for the long horizon prediction of ARDS have been validated in similar studies of mechanically ventilated patients and those with moderate hypoxia.62,63 Rule-based systems such as that developed by Herasevich et al.64,65 have been used to screen patients for ARDS by analyzing patient EHR data.66 For example, Lung Injury Prediction Score (LIPS) is a rules-based system used for predicting development of ARDS and mortality.56,67 The model allows clinicians to incorporate a series of risk factors to predict patients who will develop ARDS using clinical data at the time of presentation to the emergency depar...
Results from TrialIdentifier: No clinical trial numbers were referenced.
Results from Barzooka: We did not find any issues relating to the usage of bar graphs.
Results from JetFighter: We did not find any issues relating to colormaps.
Results from rtransparent:- Thank you for including a conflict of interest statement. Authors are encouraged to include this statement when submitting to a journal.
- Thank you for including a funding statement. Authors are encouraged to include this statement when submitting to a journal.
- No protocol registration statement was detected.
-
