COVID-19 dynamics across the US: A deep learning study of human mobility and social behavior

  1. Mohamed Aziz Bhouri
  2. Francisco Sahli Costabal
  3. Hanwen Wang
  4. Kevin Linka
  5. Mathias Peirlinck
  6. Ellen Kuhl
  7. Paris Perdikaris

Reviewed by

Read the full article See related articles

Listed in

Log in to save this article

Abstract

No abstract available

Article activity feed

  1. Version published to 10.1016/j.cma.2021.113891
  2. ScreenIT

    SciScore for 10.1101/2020.09.20.20198432: (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

    Experimental Models: Organisms/Strains
    SentencesResources
    Various multistep methods were investigated including implicit Adams-Moulton (AM) and explicit Adams-Bashford (AB) schemes with different orders.
    AB
    suggested: RRID:BDSC_203)

    Results from OddPub: Thank you for sharing your code and data.

    Results from LimitationRecognizer: We detected the following sentences addressing limitations in the study:
    Our methodology presents some limitations that open new research opportunities. First, there are pronounced differences in the testing frequency both in space and time. These disparities will affect the number of detected cases and bias the data that we use to train our model. Including the number of tests performed as part of the input to our model could improve our predictions and assess the effect of limited testing capabilities. However, these kind of data are currently not available at the county level. Secondly, we have not included a potentially large fraction of cases that remain undetected or asymptomatic. This population can have an influence on the model behavior, especially when the case numbers are relatively large. Nonetheless, estimations of the undetected population based on antibody studies are currently available only at very few locations [24]. For this reason, we decided to not explicitly model this population, as we would be simply guessing its size. Finally, our model tends to perform relatively better in counties with more cases, as we can see in Figure 4. This is intrinsically linked to the loss function that we chose to train the model. In our case, this function tends to penalize more counties with more cases. We think this behavior is desirable because it makes our predictions more accurate in regions where the pandemic is rapidly evolving. Nonetheless, it would be straightforward to adjust the loss function to emphasize other critical factors in di...

    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.

    About SciScore

    SciScore is an automated tool that is designed to assist expert reviewers by finding and presenting formulaic information scattered throughout a paper in a standard, easy to digest format. SciScore checks for the presence and correctness of RRIDs (research resource identifiers), and for rigor criteria such as sex and investigator blinding. For details on the theoretical underpinning of rigor criteria and the tools shown here, including references cited, please follow this link.

  3. Version published to 10.1101/2020.09.20.20198432v1 on medRxiv