Reference : Public Covid-19 X-ray datasets and their impact on model bias - a systematic review o...
Scientific journals : Article
Engineering, computing & technology : Computer science
Engineering, computing & technology : Computer science
Engineering, computing & technology : Multidisciplinary, general & others
Engineering, computing & technology : Multidisciplinary, general & others
Human health sciences : Radiology, nuclear medicine & imaging
Human health sciences : Radiology, nuclear medicine & imaging
Systems Biomedicine
http://hdl.handle.net/10993/46439
Public Covid-19 X-ray datasets and their impact on model bias - a systematic review of a significant problem
English
Garcia Santa Cruz, Beatriz mailto [University of Luxembourg > Luxembourg Centre for Systems Biomedicine (LCSB) > >]
Bossa, Matias Nicolas [University of Luxembourg > Luxembourg Centre for Systems Biomedicine (LCSB) > >]
Sölter, Jan mailto [University of Luxembourg > Luxembourg Centre for Systems Biomedicine (LCSB) > Interventional Neuroscience >]
Husch, Andreas mailto [University of Luxembourg > Luxembourg Centre for Systems Biomedicine (LCSB) > Interventional Neuroscience >]
Dec-2021
Medical Image Analysis
Elsevier
74
Yes (verified by ORBilu)
International
1361-8415
1361-8423
Amsterdam
Netherlands
[en] COVID-19 ; machine learning ; datasets ; X-Ray ; imaging ; review ; bias ; confounding ; confounding
[en] Computer-aided diagnosis and stratification of COVID-19 based on chest X-ray suffers from weak bias assessment and limited quality-control. Undetected bias induced by inappropriate use of datasets, and improper consideration of confounders prevents the translation of prediction models into clinical practice. By adopting established tools for model evaluation to the task of evaluating datasets, this study provides a systematic appraisal of publicly available COVID-19 chest X-ray datasets, determining their potential use and evaluating potential sources of bias. Only 9 out of more than a hundred identified datasets met at least the criteria for proper assessment of the risk of bias and could be analysed in detail. Remarkably most of the datasets utilised in 201 papers published in peer-reviewed journals, are not among these 9 datasets, thus leading to models with a high risk of bias. This raises concerns about the suitability of such models for clinical use. This systematic review highlights the limited description of datasets employed for modelling and aids researchers to select the most suitable datasets for their task.
Luxembourg Centre for Systems Biomedicine (LCSB): Systems Control (Goncalves Group) ; Luxembourg Centre for Systems Biomedicine (LCSB): Interventional Neuroscience (Hertel Group)
Fonds National de la Recherche - FnR
AICovIX
Researchers ; Professionals
http://hdl.handle.net/10993/46439
also: http://hdl.handle.net/10993/48108
10.1016/j.media.2021.102225
https://www.sciencedirect.com/science/article/pii/S136184152100270X
FnR ; FNR14702831 > Andreas Husch > AICovIX > Ai Based Diagnosis Of Covid-19 From Ct/X-ray Imaging > 01/06/2020 > 30/11/2020 > 2020

File(s) associated to this reference

Fulltext file(s):

FileCommentaryVersionSizeAccess
Open access
1-s2.0-S136184152100270X-main.pdfPublisher postprint2.25 MBView/Open

Bookmark and Share SFX Query

All documents in ORBilu are protected by a user license.