Development and Prospective Validation of a Deep Learning Algorithm for Predicting Need for Mechanical Ventilation
This article has been Reviewed by the following groups
Listed in
- Evaluated articles (ScreenIT)
Abstract
Article activity feed
-
-
SciScore for 10.1101/2020.05.30.20118109: (What is this?)
Please note, not all rigor criteria are appropriate for all manuscripts.
Table 1: Rigor
NIH rigor criteria are not applicable to paper type.Table 2: Resources
Software and Algorithms Sentences Resources VenNet was implemented in Tensorflow, version 1.12.0, machine learning frameworks for Python, version 2.7 (Python Software Foundation). Tensorflowsuggested: (tensorflow, RRID:SCR_016345)Pythonsuggested: (IPython, RRID:SCR_001658)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:Despite its many strengths, this study includes a number of limitations. First, we defined the …
SciScore for 10.1101/2020.05.30.20118109: (What is this?)
Please note, not all rigor criteria are appropriate for all manuscripts.
Table 1: Rigor
NIH rigor criteria are not applicable to paper type.Table 2: Resources
Software and Algorithms Sentences Resources VenNet was implemented in Tensorflow, version 1.12.0, machine learning frameworks for Python, version 2.7 (Python Software Foundation). Tensorflowsuggested: (tensorflow, RRID:SCR_016345)Pythonsuggested: (IPython, RRID:SCR_001658)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:Despite its many strengths, this study includes a number of limitations. First, we defined the need for mechanical ventilation in our EHR database based on the presence of PEEP and FiO2 measurements. We believe that this definition is robust based on considerable experience, but acknowledge that some mis-labeling could occur in any EHR based criteria. Nonetheless, we view such misclassification as random and do not expect any potential misclassifications would artificially improve our model’s performance. Second, more generally the proposed algorithm makes use of EHR data that was not originally designed for the analysis performed in our study. However, the superior performance of our algorithm, even in the presence of missing data, confirms its utility in a real-world clinical setting. Third, the COVID-19 pandemic has led to many changes in usual care including potentially earlier intubation, avoidance of high flow nasal cannula, and avoidance of non-invasive ventilation, among others. Thus, one could argue that the need for intubation of these patients may be driven by factors unique to this epidemic. However, our model was trained and validated with historical data from major academic centers prior to COVID-19. Thus, the high observed AUCs speak to the robustness of the model, even in the face of rapid changes in practice patterns. Fourth, one could argue that the outcome of intubation and need for mechanical ventilation is somewhat subjective and could be a function of lo...
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.
-