Serviceeinschränkungen vom 12.-22.02.2026 - weitere Infos auf der UB-Homepage
  • Automatic classification of es...

Treffer: Automatic classification of esophageal lesions in endoscopic images using a convolutional neural network.

Title:
Automatic classification of esophageal lesions in endoscopic images using a convolutional neural network.
Authors:
Liu G; Department of Geriatric Gerontology, The First Affiliated Hospital of Nanjing Medical University, Nanjing 210029, China., Hua J; Department of Gastroenterology, The First Affiliated Hospital of Nanjing Medical University, Nanjing 210029, China., Wu Z; Laboratory of Image Science and Technology, School of Computer Science and Engineering, Southeast University, Nanjing 211102, China.; The Key Laboratory of Computer Network and Information Integration (Southeast University), Ministry of Education, Nanjing 211102, China., Meng T; Laboratory of Image Science and Technology, School of Computer Science and Engineering, Southeast University, Nanjing 211102, China.; The Key Laboratory of Computer Network and Information Integration (Southeast University), Ministry of Education, Nanjing 211102, China., Sun M; Department of Geriatric Gerontology, The First Affiliated Hospital of Nanjing Medical University, Nanjing 210029, China., Huang P; Department of Geriatric Gerontology, The First Affiliated Hospital of Nanjing Medical University, Nanjing 210029, China., He X; Department of Geriatric Gerontology, The First Affiliated Hospital of Nanjing Medical University, Nanjing 210029, China., Sun W; Department of Geriatric Gerontology, The First Affiliated Hospital of Nanjing Medical University, Nanjing 210029, China., Li X; Department of Gastroenterology, The First Affiliated Hospital of Nanjing Medical University, Nanjing 210029, China., Chen Y; Laboratory of Image Science and Technology, School of Computer Science and Engineering, Southeast University, Nanjing 211102, China.; The Key Laboratory of Computer Network and Information Integration (Southeast University), Ministry of Education, Nanjing 211102, China.; Centre de Recherche en Information Biomedicale Sino-Francais (LIA CRIBs), Rennes, France.
Source:
Annals of translational medicine [Ann Transl Med] 2020 Apr; Vol. 8 (7), pp. 486.
Publication Type:
Journal Article
Language:
English
Journal Info:
Publisher: AME Publishing Company Country of Publication: China NLM ID: 101617978 Publication Model: Print Cited Medium: Print ISSN: 2305-5839 (Print) Linking ISSN: 23055839 NLM ISO Abbreviation: Ann Transl Med Subsets: PubMed not MEDLINE
Imprint Name(s):
Original Publication: [Hong Kong] : AME Publishing Company
References:
Ann Surg Oncol. 2010 Mar;17(3):784-90. (PMID: 19953333)
South Med J. 2014 Apr;107(4):225-8. (PMID: 24937515)
Nature. 2015 May 28;521(7553):436-44. (PMID: 26017442)
United European Gastroenterol J. 2019 Mar;7(2):189-198. (PMID: 31080603)
Oncology. 2017;93 Suppl 1:30-34. (PMID: 29258081)
Korean J Intern Med. 2016 Mar;31(2):253-9. (PMID: 26864297)
Opt Express. 2016 Sep 5;24(18):20799-807. (PMID: 27607683)
Clin Gastroenterol Hepatol. 2018 Oct;16(10):1585-1592. (PMID: 29174712)
Endosc Int Open. 2014 Jun;2(2):E46-50. (PMID: 26135259)
EBioMedicine. 2017 Nov;25:106-111. (PMID: 29056541)
CA Cancer J Clin. 2018 Nov;68(6):394-424. (PMID: 30207593)
Insights Imaging. 2018 Aug;9(4):611-629. (PMID: 29934920)
IEEE J Biomed Health Inform. 2017 Jan;21(1):41-47. (PMID: 28114040)
Gastrointest Endosc. 2019 Jan;89(1):25-32. (PMID: 30120958)
Gut. 2019 Jan;68(1):94-100. (PMID: 29066576)
Endoscopy. 2014 Jul;46(7):553-60. (PMID: 24971624)
Gastroenterology. 2016 Mar;150(3):599-607.e7; quiz e14-5. (PMID: 26619962)
Radiol Phys Technol. 2017 Sep;10(3):257-273. (PMID: 28689314)
Gastric Cancer. 2018 Jul;21(4):653-660. (PMID: 29335825)
J Clin Med. 2019 Jun 13;8(6):. (PMID: 31200519)
Annu Int Conf IEEE Eng Med Biol Soc. 2018 Jul;2018:4138-4141. (PMID: 30441266)
IEEE Trans Med Imaging. 2016 May;35(5):1285-98. (PMID: 26886976)
Contributed Indexing:
Keywords: Esophageal cancer (EC); convolutional neural network (CNN); deep learning; endoscopic diagnosis
Entry Date(s):
Date Created: 20200513 Latest Revision: 20220830
Update Code:
20250114
PubMed Central ID:
PMC7210177
DOI:
10.21037/atm.2020.03.24
PMID:
32395530
Database:
MEDLINE

Weitere Informationen

Background: Using deep learning techniques in image analysis is a dynamically emerging field. This study aims to use a convolutional neural network (CNN), a deep learning approach, to automatically classify esophageal cancer (EC) and distinguish it from premalignant lesions.
Methods: A total of 1,272 white-light images were adopted from 748 subjects, including normal cases, premalignant lesions, and cancerous lesions; 1,017 images were used to train the CNN, and another 255 images were examined to evaluate the CNN architecture. Our proposed CNN structure consists of two subnetworks (O-stream and P-stream). The original images were used as the inputs of the O-stream to extract the color and global features, and the pre-processed esophageal images were used as the inputs of the P-stream to extract the texture and detail features.
Results: The CNN system we developed achieved an accuracy of 85.83%, a sensitivity of 94.23%, and a specificity of 94.67% after the fusion of the 2 streams was accomplished. The classification accuracy of normal esophagus, premalignant lesion, and EC were 94.23%, 82.5%, and 77.14%, respectively, which shows a better performance than the Local Binary Patterns (LBP) + Support Vector Machine (SVM) and Histogram of Gradient (HOG) + SVM methods. A total of 8 of the 35 (22.85%) EC lesions were categorized as premalignant lesions because of their slightly reddish and flat lesions.
Conclusions: The CNN system, with 2 streams, demonstrated high sensitivity and specificity with the endoscopic images. It obtained better detection performance than the currently used methods based on the same datasets and has great application prospects in assisting endoscopists to distinguish esophageal lesion subclasses.
(2020 Annals of Translational Medicine. All rights reserved.)

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at http://dx.doi.org/10.21037/atm.2020.03.24). The authors have no conflicts of interest to declare.