Recurrent neural network models (CovRNN) for predicting outcomes of patients with COVID-19 on admission to hospital: model development and validation using electronic health record data
This article has been Reviewed by the following groups
Listed in
- Evaluated articles (ScreenIT)
Abstract
No abstract available
Article activity feed
-
-
SciScore for 10.1101/2021.09.27.21264121: (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 CovRNN also consumes the time difference between visits for a better temporal representation of patient history, which is known to slightly improve the prediction accuracy.18,19 For binary classification tasks, we compared CovRNN against traditional machine learning algorithms, such as logistic regression (LR)20 and light gradient boost machine (LGBM). CovRNNsuggested: NoneResults from OddPub: Thank you for sharing your code and data.
Results from LimitationRecognizer: We detected the following sentences addressing limitations in the study:Our study has several limitations. First, our …
SciScore for 10.1101/2021.09.27.21264121: (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 CovRNN also consumes the time difference between visits for a better temporal representation of patient history, which is known to slightly improve the prediction accuracy.18,19 For binary classification tasks, we compared CovRNN against traditional machine learning algorithms, such as logistic regression (LR)20 and light gradient boost machine (LGBM). CovRNNsuggested: NoneResults from OddPub: Thank you for sharing your code and data.
Results from LimitationRecognizer: We detected the following sentences addressing limitations in the study:Our study has several limitations. First, our data analysis includes only retrospective data. Despite our efforts to avoid potential bias by separating training, validation, and test datasets as well as external validation on a different data source, potential biases are inevitable. A prospective validation study is warranted, ideally, in hospitals that did not participate in data sharing with the database that we used to secure the validation of transferability. Second, our model focused only on the admission data to predict in-hospital clinical outcomes. It is possible to use multiple time points during the hospital stay to update models to achieve “real-time” predictions. Because minimal data preprocessing is required, our model can be easily modified to use different data points to predict future clinical outcomes. Third, we performed only a preliminary evaluation for the model predictions explanations, whereby we extracted data from 20 random sample patients and presented their predicted risk scores as well as the contribution score assigned for each medical event and asked infectious disease specialists to evaluate its relevance. Although we acknowledge that this is not a rigorous evaluation method, it demonstrated that our proposed model provides the tool that allows model transparency and helps to engage clinicians and facilitate their judgment on the model predictions. Future work is warranted to systematically evaluate the model’s transparency. Fourth, the dynamics ...
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.
Results from scite Reference Check: We found no unreliable references.
-
