AI for radiographic COVID-19 detection selects shortcuts over signal

  1. Alex J. DeGrave
  2. Joseph D. Janizek
  3. Su-In Lee

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 10.1038/s42256-021-00338-7
  2. Version published to 10.1101/2020.09.13.20193565v2 on medRxiv
  3. ScreenIT

    SciScore for 10.1101/2020.09.13.20193565: (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
    SentencesResources
    All models were trained using the PyTorch software library40, version 1.4, on NVIDIA RTX 2080 graphics processing units and required approximately 5 hours of training time per replicate.
    PyTorch
    suggested: (PyTorch, RRID:SCR_018536)
    Weal soaim tolearnthein ver setransf ormation,F :↦. Sin cegener ativea dversari al netw o rkshaveprevio uslybe enshown tobe eff ectiveforthe interpretati on ofneura lne tworks,w elearn thesetw ot ransformati onsu singtheCycle GAN app roac h17, 18.
    CycleGAN
    suggested: None
    Our CycleGAN networks were implemented in Python 3.7 using the PyTorch software library and an open-source implementation of the CycleGAN approach43.
    Python
    suggested: (IPython, RRID:SCR_001658)

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

    Results from LimitationRecognizer: An explicit section about the limitations of the techniques employed in this study was not found. We encourage authors to address study limitations.

    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.

  4. Version published to 10.1101/2020.09.13.20193565v1 on medRxiv