Common pitfalls and recommendations for using machine learning to detect and prognosticate for COVID-19 using chest radiographs and CT scans

Show simple item record

dc.contributor University of Helsinki, Department of Physics en
dc.contributor University of Helsinki, Research Program in Systems Oncology en
dc.contributor University of Helsinki, Research Program in Systems Oncology en
dc.contributor.author AIX-COVNET
dc.contributor.author Roberts, Michael
dc.contributor.author Gozaliasl, Ghassem
dc.contributor.author Tang, Jing
dc.contributor.author Shadbahr, Tolou
dc.date.accessioned 2021-05-04T11:45:01Z
dc.date.available 2021-05-04T11:45:01Z
dc.date.issued 2021-03
dc.identifier.citation AIX-COVNET , Roberts , M , Gozaliasl , G , Tang , J & Shadbahr , T 2021 , ' Common pitfalls and recommendations for using machine learning to detect and prognosticate for COVID-19 using chest radiographs and CT scans ' , Nature Machine Intelligence , vol. 3 , no. 3 , pp. 199-217 . https://doi.org/10.1038/s42256-021-00307-0 en
dc.identifier.issn 2522-5839
dc.identifier.other PURE: 161243116
dc.identifier.other PURE UUID: bad31bde-56c2-4f3a-b302-63aa8c9f13ae
dc.identifier.other RIS: urn:CFB1CF62133D384D3D6873F6EFE64FF4
dc.identifier.other RIS: Roberts2021
dc.identifier.other WOS: 000630122800002
dc.identifier.other ORCID: /0000-0001-7480-7710/work/93303056
dc.identifier.uri http://hdl.handle.net/10138/329624
dc.description.abstract Many machine learning-based approaches have been developed for the prognosis and diagnosis of COVID-19 from medical images and this Analysis identifies over 2,200 relevant published papers and preprints in this area. After initial screening, 62 studies are analysed and the authors find they all have methodological flaws standing in the way of clinical utility. The authors have several recommendations to address these issues. Machine learning methods offer great promise for fast and accurate detection and prognostication of coronavirus disease 2019 (COVID-19) from standard-of-care chest radiographs (CXR) and chest computed tomography (CT) images. Many articles have been published in 2020 describing new machine learning-based models for both of these tasks, but it is unclear which are of potential clinical utility. In this systematic review, we consider all published papers and preprints, for the period from 1 January 2020 to 3 October 2020, which describe new machine learning models for the diagnosis or prognosis of COVID-19 from CXR or CT images. All manuscripts uploaded to bioRxiv, medRxiv and arXiv along with all entries in EMBASE and MEDLINE in this timeframe are considered. Our search identified 2,212 studies, of which 415 were included after initial screening and, after quality screening, 62 studies were included in this systematic review. Our review finds that none of the models identified are of potential clinical use due to methodological flaws and/or underlying biases. This is a major weakness, given the urgency with which validated COVID-19 models are needed. To address this, we give many recommendations which, if followed, will solve these issues and lead to higher-quality model development and well-documented manuscripts. en
dc.format.extent 19
dc.language.iso eng
dc.relation.ispartof Nature Machine Intelligence
dc.rights en
dc.subject 3122 Cancers en
dc.subject 111 Mathematics en
dc.subject 113 Computer and information sciences en
dc.subject PREDICTION en
dc.subject RADIOMICS en
dc.subject VALIDATION en
dc.subject IMAGES en
dc.subject RISK en
dc.subject TOOL en
dc.title Common pitfalls and recommendations for using machine learning to detect and prognosticate for COVID-19 using chest radiographs and CT scans en
dc.type Article
dc.description.version Peer reviewed
dc.identifier.doi https://doi.org/10.1038/s42256-021-00307-0
dc.type.uri info:eu-repo/semantics/other
dc.type.uri info:eu-repo/semantics/publishedVersion
dc.contributor.pbl
dc.contributor.pbl
dc.contributor.pbl

Files in this item

Total number of downloads: Loading...

Files Size Format View
s42256_021_00307_0.pdf 1.757Mb PDF View/Open

This item appears in the following Collection(s)

Show simple item record