Rapid triage for COVID-19 using routine clinical data for patients attending hospital: development and prospective validation of an artificial intelligence screening test

  1. Andrew A S Soltan
  2. Samaneh Kouchaki
  3. Tingting Zhu
  4. Dani Kiyasseh
  5. Thomas Taylor
  6. Zaamin B Hussain
  7. Tim Peto
  8. Andrew J Brent
  9. David W Eyre
  10. David A Clifton

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/s2589-7500(20)30274-0
  2. ScreenIT

    SciScore for 10.1101/2020.07.07.20148361: (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
    For each presentation, data extracted included admission blood tests, blood gas testing, vital signs, results of SARS-CoV-2 RT-PCR assays (Public Health England designed RdRp and Abbott Architect [Abbott, Maidenhead, UK]) of nasopharyngeal swabs, and PCR for influenza and other respiratory viruses.
    Abbott Architect
    suggested: (Abbott ARCHITECT i1000sr System, RRID:SCR_019328)
    Abbott
    suggested: (Abbott, RRID:SCR_010477)

    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:
    Limitations of the gold-standard PCR test for COVID-19 have challenged healthcare systems across the world. There remains an urgent clinical need for rapid and accurate testing on arrival to hospitals, with the current test limited by prolonged turnaround times18, shortages of specialist equipment and operators, and relatively low sensitivity8. In this study, we develop and assess two Artificial Intelligence (AI) driven screening tools for in-hospital COVID-19 screening, in the clinical context intended for use. Our Emergency Department and Admission models effectively identify patients with COVID-19 amongst all patients presenting and admitted to hospital, using data typically available within the first hour of presentation (AUROC 0.939 & 0.940). On validation using appropriate prospective cohorts of all patients presenting or admitted to a large UK teaching centre group between 20th April and 6th May 2020 (n=3,326 & 1,715), our models achieve high accuracies (92.3% and 92.5%) and negative predictive values (97.6% and 97.7%). A sensitivity analysis to account for uncertainty in negative PCR results improves apparent accuracy (95.1% and 94.1%) and NPV (99.0% and 98.5%). Simulation on test-sets with varying prevalences of COVID-19 shows that our models achieve clinically useful negative predictive values (>0.99) at low prevalences (<5%), supporting safe exclusion of the disease. At higher prevalences (>25%), our models can be configured to meet clinical needs for higher positi...

    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 10.1101/2020.07.07.20148361v1 on medRxiv