Histopathological Classification of Canine Cutaneous Round Cell Tumors Using Deep Learning: A Multi-Center Study.
computer-aided image analysis
cutaneous round cell tumors
deep learning
digital pathology
dog
mast cell tumors
Journal
Frontiers in veterinary science
ISSN: 2297-1769
Titre abrégé: Front Vet Sci
Pays: Switzerland
ID NLM: 101666658
Informations de publication
Date de publication:
2021
2021
Historique:
received:
12
12
2020
accepted:
08
03
2021
entrez:
19
4
2021
pubmed:
20
4
2021
medline:
20
4
2021
Statut:
epublish
Résumé
Canine cutaneous round cell tumors (RCT) represent one of the routine diagnostic challenges for veterinary pathologists. Computer-aided approaches are developed to overcome these restrictions and to increase accuracy and consistency of diagnosis. These systems are also of high benefit reducing errors when a large number of cases are screened daily. In this study we describe ARCTA (Automated Round Cell Tumors Assessment), a fully automated algorithm for cutaneous RCT classification and mast cell tumors grading in canine histopathological images. ARCTA employs a deep learning strategy and was developed on 416 RCT images and 213 mast cell tumors images. In the test set, our algorithm exhibited an excellent classification performance in both RCT classification (accuracy: 91.66%) and mast cell tumors grading (accuracy: 100%). Misdiagnoses were encountered for histiocytomas in the train set and for melanomas in the test set. For mast cell tumors the reduction of a grade was observed in the train set, but not in the test set. To the best of our knowledge, the proposed model is the first fully automated algorithm in histological images specifically developed for veterinary medicine. Being very fast (average computational time 2.63 s), this algorithm paves the way for an automated and effective evaluation of canine tumors.
Identifiants
pubmed: 33869320
doi: 10.3389/fvets.2021.640944
pmc: PMC8044886
doi:
Types de publication
Journal Article
Langues
eng
Pagination
640944Informations de copyright
Copyright © 2021 Salvi, Molinari, Iussich, Muscatello, Pazzini, Benali, Banco, Abramo, De Maria and Aresu.
Déclaration de conflit d'intérêts
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
Références
Biomed Eng Online. 2018 Jun 20;17(1):89
pubmed: 29925379
ILAR J. 2018 Dec 1;59(1):66-79
pubmed: 30535284
Vet Res Commun. 2015 Mar;39(1):7-17
pubmed: 25563490
Comput Biol Med. 2021 Jan;128:104129
pubmed: 33254082
J Pathol Inform. 2016 Jul 26;7:29
pubmed: 27563488
Comput Methods Programs Biomed. 2020 Sep;193:105506
pubmed: 32353672
Proc IEEE Comput Soc Conf Comput Vis Pattern Recognit. 2016 Jun-Jul;2016:2424-2433
pubmed: 27795661
Med Image Anal. 2016 Oct;33:170-175
pubmed: 27423409
Micron. 2018 Nov;114:42-61
pubmed: 30096632
Folia Histochem Cytobiol. 2019;57(3):146-154
pubmed: 31553052
Vet Pathol. 2018 Jul;55(4):490-500
pubmed: 29402206
Compend Contin Educ Vet. 2008 Apr;30(4):202-4, 208-16; quiz 216-17
pubmed: 18576276
Genes (Basel). 2019 Jun 30;10(7):
pubmed: 31262050
Vet Pathol. 2002 Nov;39(6):651-78
pubmed: 12450197
Vet Pathol. 2011 Jan;48(1):32-40
pubmed: 21078882
Nat Rev Clin Oncol. 2019 Nov;16(11):703-715
pubmed: 31399699
Vet Pathol. 2011 Jan;48(1):147-55
pubmed: 21062911
Vet Pathol. 2005 Jul;42(4):437-45
pubmed: 16006603
Am J Pathol. 2019 Sep;189(9):1686-1698
pubmed: 31199919