Automated diagnosis of ear disease using ensemble deep learning with a big otoendoscopy image database.
Convolutional neural network
Deep learning
Ear disease
Ensemble learning
Otoendoscopy
Tympanic membrane
Journal
EBioMedicine
ISSN: 2352-3964
Titre abrégé: EBioMedicine
Pays: Netherlands
ID NLM: 101647039
Informations de publication
Date de publication:
Jul 2019
Jul 2019
Historique:
received:
21
05
2019
revised:
19
06
2019
accepted:
25
06
2019
pubmed:
6
7
2019
medline:
18
12
2019
entrez:
6
7
2019
Statut:
ppublish
Résumé
Ear and mastoid disease can easily be treated by early detection and appropriate medical care. However, short of specialists and relatively low diagnostic accuracy calls for a new way of diagnostic strategy, in which deep learning may play a significant role. The current study presents a machine learning model to automatically diagnose ear disease using a large database of otoendoscopic images acquired in the clinical environment. Total 10,544 otoendoscopic images were used to train nine public convolution-based deep neural networks to classify eardrum and external auditory canal features into six categories of ear diseases, covering most ear diseases (Normal, Attic retraction, Tympanic perforation, Otitis externa±myringitis, Tumor). After evaluating several optimization schemes, two best-performing models were selected to compose an ensemble classifier, by combining classification scores of each classifier. According to accuracy and training time, transfer learning models based on Inception-V3 and ResNet101 were chosen and the ensemble classifier using the two models yielded a significant improvement over each model, the accuracy of which is in average 93·67% for the 5-folds cross-validation. Considering substantial data-size dependency of classifier performance in the transfer learning, evaluated in this study, the high accuracy in the current model is attributable to the large database. The current study is unprecedented in terms of both disease diversity and diagnostic accuracy, which is compatible or even better than an average otolaryngologist. The classifier was trained with data in a various acquisition condition, which is suitable for the practical environment. This study shows the usefulness of utilizing a deep learning model in the early detection and treatment of ear disease in the clinical situation. FUND: This research was supported by Brain Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Science and ICT(NRF-2017M3C7A1049051).
Sections du résumé
BACKGROUND
BACKGROUND
Ear and mastoid disease can easily be treated by early detection and appropriate medical care. However, short of specialists and relatively low diagnostic accuracy calls for a new way of diagnostic strategy, in which deep learning may play a significant role. The current study presents a machine learning model to automatically diagnose ear disease using a large database of otoendoscopic images acquired in the clinical environment.
METHODS
METHODS
Total 10,544 otoendoscopic images were used to train nine public convolution-based deep neural networks to classify eardrum and external auditory canal features into six categories of ear diseases, covering most ear diseases (Normal, Attic retraction, Tympanic perforation, Otitis externa±myringitis, Tumor). After evaluating several optimization schemes, two best-performing models were selected to compose an ensemble classifier, by combining classification scores of each classifier.
FINDINGS
RESULTS
According to accuracy and training time, transfer learning models based on Inception-V3 and ResNet101 were chosen and the ensemble classifier using the two models yielded a significant improvement over each model, the accuracy of which is in average 93·67% for the 5-folds cross-validation. Considering substantial data-size dependency of classifier performance in the transfer learning, evaluated in this study, the high accuracy in the current model is attributable to the large database.
INTERPRETATION
CONCLUSIONS
The current study is unprecedented in terms of both disease diversity and diagnostic accuracy, which is compatible or even better than an average otolaryngologist. The classifier was trained with data in a various acquisition condition, which is suitable for the practical environment. This study shows the usefulness of utilizing a deep learning model in the early detection and treatment of ear disease in the clinical situation. FUND: This research was supported by Brain Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Science and ICT(NRF-2017M3C7A1049051).
Identifiants
pubmed: 31272902
pii: S2352-3964(19)30431-1
doi: 10.1016/j.ebiom.2019.06.050
pmc: PMC6642402
pii:
doi:
Types de publication
Journal Article
Langues
eng
Sous-ensembles de citation
IM
Pagination
606-614Informations de copyright
Copyright © 2019 The Authors. Published by Elsevier B.V. All rights reserved.
Références
Arch Pediatr Adolesc Med. 2001 Oct;155(10):1137-42
pubmed: 11576009
Ann Otol Rhinol Laryngol. 1976 Mar-Apr;85(2 Suppl 25 Pt 2):66-72
pubmed: 1267370
Fam Pract. 2003 Oct;20(5):524-7
pubmed: 14507792
Int J Pediatr Otorhinolaryngol. 2005 Mar;69(3):361-6
pubmed: 15733595
Telemed J E Health. 2013 Apr;19(4):252-8
pubmed: 23384332
IEEE Trans Med Imaging. 2016 May;35(5):1285-98
pubmed: 26886976
EBioMedicine. 2016 Feb 08;5:156-60
pubmed: 27077122
JAMA. 2016 Dec 13;316(22):2402-2410
pubmed: 27898976
Biomed Opt Express. 2017 Jan 04;8(2):579-592
pubmed: 28270969
J Telemed Telecare. 2018 Aug;24(7):453-459
pubmed: 28480781
Cell. 2018 Feb 22;172(5):1122-1131.e9
pubmed: 29474911
Ophthalmology. 2018 Sep;125(9):1410-1420
pubmed: 29653860
Ophthalmology. 2018 Aug;125(8):1207-1208
pubmed: 30032794
Am J Ophthalmol. 2019 Feb;198:136-145
pubmed: 30316669
Comput Methods Programs Biomed. 2018 Oct;165:69-76
pubmed: 30337082
Otolaryngol Head Neck Surg. 2019 Mar;160(3):447-456
pubmed: 30396324
Adv Otorhinolaryngol. 1988;40:110-7
pubmed: 3389223
Acta Otolaryngol. 1980 May-Jun;89(5-6):479-86
pubmed: 7192477