A Web-Based Deep Learning Model for Automated Diagnosis of Otoscopic Images.
Journal
Otology & neurotology : official publication of the American Otological Society, American Neurotology Society [and] European Academy of Otology and Neurotology
ISSN: 1537-4505
Titre abrégé: Otol Neurotol
Pays: United States
ID NLM: 100961504
Informations de publication
Date de publication:
01 10 2021
01 10 2021
Historique:
pubmed:
1
7
2021
medline:
8
10
2021
entrez:
30
6
2021
Statut:
ppublish
Résumé
To develop a multiclass-classifier deep learning model and website for distinguishing tympanic membrane (TM) pathologies based on otoscopic images. An otoscopic image database developed by utilizing publicly available online images and open databases was assessed by convolutional neural network (CNN) models including ResNet-50, Inception-V3, Inception-Resnet-V2, and MobileNetV2. Training and testing were conducted with a 75:25 breakdown. Area under the curve of receiver operating characteristics (AUC-ROC), accuracy, sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) were used to compare different CNN models' performances in classifying TM images. Our database included 400 images, organized into normal (n = 196) and abnormal classes (n = 204), including acute otitis media (n = 116), otitis externa (n = 44), chronic suppurative otitis media (n = 23), and cerumen impaction (n = 21). For binary classification between normal versus abnormal TM, the best performing model had average AUC-ROC of 0.902 (MobileNetV2), followed by 0.745 (Inception-Resnet-V2), 0.731 (ResNet-50), and 0.636 (Inception-V3). Accuracy ranged between 0.73-0.77, sensitivity 0.72-0.88, specificity 0.58-0.84, PPV 0.68-0.81, and NPV 0.73-0.83. Macro-AUC-ROC for MobileNetV2 based multiclass-classifier was 0.91, with accuracy of 66%. Binary and multiclass-classifier models based on MobileNetV2 were loaded onto a publicly accessible and user-friendly website (https://headneckml.com/tympanic). This allows the readership to upload TM images for real-time predictions using the developed algorithms. Novel CNN algorithms were developed with high AUC-ROCs for differentiating between various TM pathologies. This was further deployed as a proof-of-concept publicly accessible website for real-time predictions.
Identifiants
pubmed: 34191783
doi: 10.1097/MAO.0000000000003210
pii: 00129492-202110000-00049
pmc: PMC8448915
mid: NIHMS1694220
doi:
Types de publication
Journal Article
Research Support, N.I.H., Extramural
Langues
eng
Sous-ensembles de citation
IM
Pagination
e1382-e1388Subventions
Organisme : NCATS NIH HHS
ID : KL2 TR001416
Pays : United States
Organisme : NCATS NIH HHS
ID : TL1 TR001415
Pays : United States
Informations de copyright
Copyright © 2021, Otology & Neurotology, Inc.
Déclaration de conflit d'intérêts
The authors disclose no conflicts of interest.
Références
Davies J, Djelic L, Campisi P, Forte V, Chiodo A. Otoscopy simulation training in a classroom setting: a novel approach to teaching otoscopy to medical students. Laryngoscope 2014; 124:2594–2597.
Blomgren K, Pitkäranta A. Is it possible to diagnose acute otitis media accurately in primary health care? Fam Pract 2003; 20:524–527.
Pichichero ME, Poole MD. Assessing diagnostic accuracy and tympanocentesis skills in the management of otitis media. Arch Pediatr Adolesc Med 2001; 155:1137–1142.
Fagan JJ, Jacobs M. Survey of ENT services in Africa: need for a comprehensive intervention. Glob Health Action 2009; 2: doi:10.3402/gha.v2i0.1932.
doi: 10.3402/gha.v2i0.1932
Monasta L, Ronfani L, Marchetti F, et al. Burden of disease caused by otitis media: systematic review and global estimates. PLoS One 2012; 7:e36226.
Global Burden of Disease Study 2013 Collaborators. Global, regional, and national incidence, prevalence, and years lived with disability for 301 acute and chronic diseases and injuries in 188 countries, 1990–2013: a systematic analysis for the Global Burden of Disease Study 2013. Lancet 2015; 386:743–800.
Yamamoto Y, Tsuzuki T, Akatsuka J, et al. Automated acquisition of explainable knowledge from unannotated histopathology images. Nat Commun 2019; 10:5642.
Esteva A, Kuprel B, Novoa RA, et al. Dermatologist-level classification of skin cancer with deep neural networks. Nature 2017; 542:115–118.
De Fauw J, Ledsam JR, Romera-Paredes B, et al. Clinically applicable deep learning for diagnosis and referral in retinal disease. Nat Med 2018; 24:1342–1350.
Livingstone D, Talai AS, Chau J, Forkert ND. Building an Otoscopic screening prototype tool using deep learning. J Otolaryngol Head Neck Surg 2019; 48:66.
Cha D, Pae C, Seong SB, Choi JY, Park HJ. Automated diagnosis of ear disease using ensemble deep learning with a big otoendoscopy image database. EBioMedicine 2019; 45:606–614.
Richards JR, Gaylor KA, Pilgrim AJ. Comparison of traditional otoscope to iPhone otoscope in the pediatric ED. Am J Emerg Med 2015; 33:1089–1092.
Moshtaghi O, Sahyouni R, Haidar YM, et al. Smartphone-enabled otoscopy in neurotology/otology. Otolaryngol Head Neck Surg 2017; 156:554–558.
Coulibaly JT, Ouattara M, D’Ambrosio MV, et al. Accuracy of mobile phone and handheld light microscopy for the diagnosis of schistosomiasis and intestinal protozoa infections in Côte d’Ivoire. PLoS Negl Trop Dis 2016; 10:e0004768.
Chang J, Arbeláez P, Switz N, et al. Automated tuberculosis diagnosis using fluorescence images from a mobile microscope. Med Image Comput Comput Assist Interv 2012; 15 (Pt 3):345–352.
Başaran E, Cömert Z, Çelik Y. Convolutional neural network approach for automatic tympanic membrane detection and classification. Biomed Signal Process Control 2020; 56:101734.
Google. Google Images. 2020. Available at: https://www.google.com/imghp?hl=EN . Accessed July 2020.
He K, Zhang X, Ren S, Sun J. Deep residual learning for image recognition. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition; 2016; 770–8.
Szegedy C, Vanhoucke V, Ioffe S, Shlens J, Wojna Z. Rethinking the inception architecture for computer vision. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition; 2016; 2818–26.
Sandler M, Howard A, Zhu M, Zhmoginov A, Chen LC. Mobilenetv2: Inverted residuals and linear bottlenecks. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition; 2018; 4510–20.
Szegedy C, Ioffe S, Vanhoucke V, Alemi A. Inception-v4, inception-resnet and the impact of residual connections on learning. arXiv 2016; 160207261.
Abadi M, Agarwal A, Barham P, et al. Tensorflow: large-scale machine learning on heterogeneous distributed systems. arXiv 2016; 160304467.
Yosinski J, Clune J, Bengio Y, Lipson H. How transferable are features in deep neural networks? arXiv 2014; 1411:1792.
Zhang Z, Beck MW, Winkler DA, et al. Opening the black box of neural networks: methods for interpreting neural network models in clinical applications. Ann Transl Med 2018; 6:216.
Harmes KM, Blackwood RA, Burrows HL, Cooke JM, Harrison RV, Passamani PP. Otitis media: diagnosis and treatment. Am Fam Physician 2013; 88:435–440.
Rosenfeld RM, Schwartz SR, Cannon CR, et al. Clinical practice guideline: acute otitis externa. Otolaryngol Head Neck Surg 2014; 150: (1 Suppl): S1–S24.
Rosenfeld RM, Shin JJ, Schwartz SR, et al. Clinical practice guideline: Otitis media with effusion (update). Otolaryngol Head Neck Surg 2016; 154: (1 Suppl): S1–S41.
Schwartz SR, Magit AE, Rosenfeld RM, et al. Clinical practice guideline (update): Earwax (Cerumen Impaction). Otolaryngol Head Neck Surg 2017; 156: (1 Suppl): S1–S29.
London AJ. Artificial intelligence and black-box medical decisions: accuracy versus explainability. Hastings Cent Rep 2019; 49:15–21.