Detecting Eczema Areas in Digital Images: An Impossible Task?

AD, atopic dermatitis ICC, intraclass correlation coefficient IRR, inter-rater reliability KA, Krippendorff’s alpha ML, machine learning

Journal

JID innovations : skin science from molecules to population health
ISSN: 2667-0267
Titre abrégé: JID Innov
Pays: Netherlands
ID NLM: 101776173

Informations de publication

Date de publication:
Sep 2022
Historique:
received: 19 10 2021
revised: 28 04 2022
accepted: 02 05 2022
entrez: 12 9 2022
pubmed: 13 9 2022
medline: 13 9 2022
Statut: epublish

Résumé

Assessing the severity of atopic dermatitis (AD, or eczema) traditionally relies on a face-to-face assessment by healthcare professionals and may suffer from inter- and intra-rater variability. With the expanding role of telemedicine, several machine learning algorithms have been proposed to automatically assess AD severity from digital images. Those algorithms usually detect and then delineate (segment) AD lesions before assessing lesional severity and are trained using the data of AD areas detected by healthcare professionals. To evaluate the reliability of such data, we estimated the inter-rater reliability of AD segmentation in digital images. Four dermatologists independently segmented AD lesions in 80 digital images collected in a published clinical trial. We estimated the inter-rater reliability of the AD segmentation using the intraclass correlation coefficient at the pixel and the area levels for different resolutions of the images. The average intraclass correlation coefficient was 0.45 (

Identifiants

pubmed: 36090300
doi: 10.1016/j.xjidi.2022.100133
pii: S2667-0267(22)00041-8
pmc: PMC9460154
doi:

Types de publication

Journal Article

Langues

eng

Pagination

100133

Subventions

Organisme : Medical Research Council
ID : MC_PC_19040
Pays : United Kingdom

Informations de copyright

© 2022 The Authors.

Références

PLoS Med. 2011 Feb 15;8(2):e1000395
pubmed: 21358807
Exp Dermatol. 2018 Apr;27(4):340-357
pubmed: 29457272
Biol Rev Camb Philos Soc. 2010 Nov;85(4):935-56
pubmed: 20569253
J R Soc Interface. 2018 Apr;15(141):
pubmed: 29618526
J Chiropr Med. 2016 Jun;15(2):155-63
pubmed: 27330520
Br J Dermatol. 1996 Sep;135 Suppl 48:25-30
pubmed: 8881901
Tutor Quant Methods Psychol. 2012;8(1):23-34
pubmed: 22833776
Lancet. 2020 Aug 1;396(10247):345-360
pubmed: 32738956
Br J Dermatol. 1999 Jan;140(1):109-11
pubmed: 10215778
Annu Int Conf IEEE Eng Med Biol Soc. 2016 Aug;2016:1365-1368
pubmed: 28268579
Br J Dermatol. 1995 Dec;133(6):941-9
pubmed: 8547049
Sci Rep. 2021 Mar 15;11(1):6049
pubmed: 33723375
Med Image Anal. 2020 Oct;65:101759
pubmed: 32623277

Auteurs

Guillem Hurault (G)

Department of Bioengineering, Imperial College London, London, United Kingdom.

Kevin Pan (K)

Department of Bioengineering, Imperial College London, London, United Kingdom.

Ricardo Mokhtari (R)

Department of Bioengineering, Imperial College London, London, United Kingdom.

Bayanne Olabi (B)

Biosciences Institute, Faculty of Medical Sciences, Newcastle University, United Kingdom.

Eleanor Earp (E)

Department of Dermatology, Lauriston Building, Edinburgh, United Kingdom.

Lloyd Steele (L)

Department of Dermatology, The Royal London Hospital, Barts Health NHS Trust, London, United Kingdom.

Hywel C Williams (HC)

Biosciences Institute, Faculty of Medical Sciences, Newcastle University, United Kingdom.
Centre of Evidence-Based Dermatology, University of Nottingham, Nottingham, United Kingdom.

Reiko J Tanaka (RJ)

Department of Bioengineering, Imperial College London, London, United Kingdom.

Classifications MeSH