Classification of racehorse limb radiographs using deep convolutional neural networks.

Thoroughbred racehorse deep learning machine learning radiograph

Journal

Veterinary record open
ISSN: 2052-6113
Titre abrégé: Vet Rec Open
Pays: United States
ID NLM: 101653671

Informations de publication

Date de publication:
Jun 2023
Historique:
received: 25 05 2022
revised: 07 12 2022
accepted: 16 12 2022
entrez: 2 2 2023
pubmed: 3 2 2023
medline: 3 2 2023
Statut: epublish

Résumé

To assess the capability of deep convolutional neural networks to classify anatomical location and projection from a series of 48 standard views of racehorse limbs. Radiographs ( Top-1 accuracy of six deep learning architectures ranged from 0.737 to 0.841. Top-1 accuracy of the best deep learning architecture (ResNet-34) ranged from 0.809 to 0.878, depending on batch size. ResNet-34 (batch size = 8) achieved the highest top-1 accuracy (0.878) and the majority (91.8%) of misclassification was due to laterality error. Class activation maps indicated that joint morphology, not side markers or other non-anatomical image regions, drove the model decision. Deep convolutional neural networks can classify equine pre-import radiographs into the 48 standard views including moderate discrimination of laterality, independent of side marker presence.

Identifiants

pubmed: 36726400
doi: 10.1002/vro2.55
pii: VRO255
pmc: PMC9884469
doi:

Banques de données

figshare
['10.6084/m9.figshare.c.5921813']

Types de publication

Journal Article

Langues

eng

Pagination

e55

Informations de copyright

© 2023 The Authors. Veterinary Record Open published by John Wiley & Sons Ltd on behalf of British Veterinary Association.

Déclaration de conflit d'intérêts

The authors declare they have no conflicts of interest.

Références

Am J Orthod Dentofacial Orthop. 2021 Aug;160(2):170-192.e4
pubmed: 34103190
J Equine Vet Sci. 2020 Oct;93:103195
pubmed: 32972684
Sensors (Basel). 2021 Aug 09;21(16):
pubmed: 34450821
Aust Vet J. 2005 Jun;83(6):367-70
pubmed: 15986917
Nat Biomed Eng. 2018 Oct;2(10):719-731
pubmed: 31015651
J Clin Ultrasound. 2022 Nov;50(9):1414-1431
pubmed: 36069404
J Digit Imaging. 2019 Aug;32(4):656-664
pubmed: 31065828
Plast Surg (Oakv). 2021 May;29(2):75-80
pubmed: 34026669
Am J Vet Res. 2022 Mar 30;83(5):385-392
pubmed: 35353711
Med Image Anal. 2021 Aug;72:102125
pubmed: 34171622
Acta Orthop. 2017 Dec;88(6):581-586
pubmed: 28681679
Med Image Anal. 2017 Dec;42:60-88
pubmed: 28778026
R Soc Open Sci. 2018 Jan 24;5(1):171782
pubmed: 29410871
Am J Phys Anthropol. 1994 Jun;94(2):203-11
pubmed: 8085612
Equine Vet J. 2021 Mar;53(2):277-286
pubmed: 32654167
BMC Vet Res. 2022 Aug 31;18(1):328
pubmed: 36045350
J Digit Imaging. 2021 Feb;34(1):66-74
pubmed: 33263143
Equine Vet J. 2003 Nov;35(7):712-4
pubmed: 14649365
Eur Radiol Exp. 2018 Nov 14;2(1):36
pubmed: 30426318
Vet Pathol. 2019 Jul;56(4):512-525
pubmed: 30866728
Semin Musculoskelet Radiol. 2019 Jun;23(3):304-311
pubmed: 31163504
Equine Vet J. 2019 Jan;51(1):83-89
pubmed: 29806972

Auteurs

Raniere Gaia Costa da Silva (RG)

Department of Infectious Diseases and Public Health City University of Hong Kong Hong Kong SAR China.

Ambika Prasad Mishra (AP)

Department of Infectious Diseases and Public Health City University of Hong Kong Hong Kong SAR China.

Christopher Michael Riggs (CM)

Department of Veterinary Clinical Services Hong Kong Jockey Club Hong Kong SAR China.

Michael Doube (M)

Department of Infectious Diseases and Public Health City University of Hong Kong Hong Kong SAR China.

Classifications MeSH