Abstract
Background
Photodocumentation during endoscopy procedures is one of the indicators for endoscopy performance quality; however, this indicator is difficult to measure and audit in the endoscopy unit. Emerging artificial intelligence technology may solve this problem, which requires a large amount of material for model development. We developed a deep learning-based endoscopic anatomy classification system through convolutional neural networks with an accelerated data preparation approach.
Patients and methods
We retrospectively collected 8,041 images from esophagogastroduodenoscopy (EGD) procedures and labeled them using two experts for nine anatomical locations of the upper gastrointestinal tract. A base model for EGD image multiclass classification was first developed, and an additional 6,091 images were enrolled and classified by the base model. A total of 5,963 images were manually confirmed and added to develop the subsequent enhanced model. Additional internal and external endoscopy image datasets were used to test the model performance.
Results
The base model achieved total accuracy of 96.29%. For the enhanced model, the total accuracy was 96.64%. The overall accuracy improved with the enhanced model compared with the base model for the internal test dataset without narrowband images (93.05% vs. 91.25%, p < 0.01) or with narrowband images (92.74% vs. 90.46%, p < 0.01). The total accuracy was 92.56% of the enhanced model on the external test dataset.
Conclusions
We constructed a deep learning-based model with an accelerated approach that can be used for quality control in endoscopy units. The model was also validated with both internal and external datasets with high accuracy.
Similar content being viewed by others
References
Attardo S, Chandrasekar VT, Spadaccini M, Maselli R, Patel HK, Desai M et al (2020) Artificial intelligence technologies for the detection of colorectal lesions: the future is now. World J Gastroenterol 26:5606–5616
Luo H, Xu G, Li C, He L, Luo L, Wang Z et al (2019) Real-time artificial intelligence for detection of upper gastrointestinal cancer by endoscopy: a multicentre, case-control, diagnostic study. Lancet Oncol 20:1645–1654
Aoki T, Yamada A, Aoyama K, Saito H, Tsuboi A, Nakada A et al (2019) Automatic detection of erosions and ulcerations in wireless capsule endoscopy images based on a deep convolutional neural network. Gastrointest Endosc 89:357–63 e2
Pimenta-Melo AR, Monteiro-Soares M, Libanio D, Dinis-Ribeiro M (2016) Missing rate for gastric cancer during upper gastrointestinal endoscopy: a systematic review and meta-analysis. Eur J Gastroenterol Hepatol 28:1041–1049
Park JM, Huo SM, Lee HH, Lee BI, Song HJ, Choi MG (2017) Longer observation time increases proportion of neoplasms detected by esophagogastroduodenoscopy. Gastroenterology 153:460–9 e1
Park JM, Lim CH, Cho YK, Lee BI, Cho YS, Song HJ et al (2019) The effect of photo-documentation of the ampulla on neoplasm detection rate during esophagogastroduodenoscopy. Endoscopy 51:115–124
Marques S, Bispo M, Pimentel-Nunes P, Chagas C, Dinis-Ribeiro M (2017) Image documentation in gastrointestinal endoscopy: review of recommendations. GE Port J Gastroenterol 24:269–274
Rey JF, Lambert R, Committee EQA (2001) ESGE recommendations for quality control in gastrointestinal endoscopy: guidelines for image documentation in upper and lower GI endoscopy. Endoscopy 33:901–903
Emura F, Sharma P, Arantes V, Cerisoli C, Parra-Blanco A, Sumiyama K et al (2020) Principles and practice to facilitate complete photodocumentation of the upper gastrointestinal tract: World Endoscopy Organization position statement. Dig Endosc 32:168–179
Beg S, Ragunath K, Wyman A, Banks M, Trudgill N, Pritchard DM et al (2017) Quality standards in upper gastrointestinal endoscopy: a position statement of the British Society of Gastroenterology (BSG) and Association of Upper Gastrointestinal Surgeons of Great Britain and Ireland (AUGIS). Gut 66:1886–1899
Bisschops R, Rutter MD, Areia M, Spada C, Domagk D, Kaminski MF et al (2021) Overcoming the barriers to dissemination and implementation of quality measures for gastrointestinal endoscopy: European Society of Gastrointestinal Endoscopy (ESGE) and United European Gastroenterology (UEG) position statement. Endoscopy. https://doi.org/10.1177/2050640620981366
Borgli H, Thambawita V, Smedsrud PH, Hicks S, Jha D, Eskeland SL et al (2020) HyperKvasir, a comprehensive multi-class image and video dataset for gastrointestinal endoscopy. Sci Data 7:283
Zhang H, Wu C, Zhang Z, Zhu Y, Lin H, Zhang Z, et al. ResNeSt: Split-Attention Networks. 2020: arXiv:2004.08955.
He K, Zhang X, Ren S, Sun J. (2016) Deep residual learning for image recognition. 2016 IEEE Conference on computer vision and pattern recognition (CVPR). p. 770–778
Deng J, Dong W, Socher R, Li L, Kai L, Li F-F. (2009) ImageNet: a large-scale hierarchical image database. 2009 IEEE Conference on computer vision and pattern recognition. p. 248–255
Lin T-Y, Goyal P, Girshick R, He K, Dollár P. Focal loss for dense object detection. 2017: arXiv:1708.02002.
Dietterich TG (1998) Approximate statistical tests for comparing supervised classification learning algorithms. Neural Comput 10:1895–1923
Selvaraju RR, Cogswell M, Das A, Vedantam R, Parikh D, Batra D. (2017) Grad-CAM: visual explanations from deep networks via gradient-based localization. 2017 IEEE International conference on computer vision (ICCV). p 618–626
Barkun AN, Almadi M, Kuipers EJ, Laine L, Sung J, Tse F et al (2019) Management of nonvariceal upper gastrointestinal bleeding: guideline recommendations from the international consensus group. Ann Intern Med 171:805–822
Penny HA, Kurien M, Wong E, Ahmed R, Ejenavi E, Lau M et al (2016) Changing trends in the UK management of upper GI bleeding: is there evidence of reduced UK training experience? Frontline Gastroenterol 7:67–72
Cohen J, Pike IM (2015) Defining and measuring quality in endoscopy. Gastrointest Endosc 81:1–2
Bisschops R, Areia M, Coron E, Dobru D, Kaskas B, Kuvaev R et al (2016) Performance measures for upper gastrointestinal endoscopy: a European Society of Gastrointestinal Endoscopy (ESGE) quality improvement initiative. Endoscopy 48:843–864
Rutter MD, Senore C, Bisschops R, Domagk D, Valori R, Kaminski MF et al (2016) The European society of gastrointestinal endoscopy quality improvement initiative: developing performance measures. Endoscopy 48:81–89
Ang TL, Carneiro G (2021) Artificial intelligence in gastrointestinal endoscopy. J Gastroenterol Hepatol 36:5–6
Takiyama H, Ozawa T, Ishihara S, Fujishiro M, Shichijo S, Nomura S et al (2018) Automatic anatomical classification of esophagogastroduodenoscopy images using deep convolutional neural networks. Sci Rep 8:7497
Xu Z, Tao Y, Wenfang Z, Ne L, Zhengxing H, Jiquan L et al (2019) Upper gastrointestinal anatomy detection with multi-task convolutional neural networks. Healthc Technol Lett 6:176–180
He Q, Bano S, Ahmad OF, Yang B, Chen X, Valdastri P et al (2020) Deep learning-based anatomical site classification for upper gastrointestinal endoscopy. Int J Comput Assist Radiol Surg 15:1085–1094
Chen CH, Lee YW, Huang YS, Lan WR, Chang RF, Tu CY et al (2019) Computer-aided diagnosis of endobronchial ultrasound images using convolutional neural network. Comput Methods Programs Biomed 177:175–182
Chang RF, Lee CC, Lo CM (2019) Quantitative diagnosis of rotator cuff tears based on sonographic pattern recognition. PLoS ONE 14:e0212741
Choi SJ, Khan MA, Choi HS, Choo J, Lee JM, Kwon S et al (2021) Development of artificial intelligence system for quality control of photo documentation in esophagogastroduodenoscopy. Surg Endosc. https://doi.org/10.1007/s00464-020-08236-6
Yen HH, Wu PY, Su PY, Yang CW, Chen YY, Chen MF et al (2021) Performance comparison of the deep learning and the human endoscopist for bleeding peptic ulcer disease. J Med Biol Eng. https://doi.org/10.1007/s40846-021-00608-0
Funding
The authors thank the Ministry of Science and Technology of Taiwan (MOST 109-2634-F-002-026 and MOST 110-2634-F-002-009) and Changhua Christian Hospital (109-CCH-IRP-008, 110-CCH-IRP-020) for financial support.
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Disclosures
Yuan-Yen Chang, Pai-Chi Li, Ruey-Feng Chang, Chih-Da Yao, Yang-Yuan Chen, Wen-Yen Chang and Hsu-Heng Yen have no conflict of interest or financial ties to disclose.
Ethical approval
The present study was approved by the Ethical Review Committee of Changhua Christian Hospital. All procedures were performed in accordance with the ethical standards laid down in the 1964 Declaration of Helsinki and its later amendments.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Chang, YY., Li, PC., Chang, RF. et al. Deep learning-based endoscopic anatomy classification: an accelerated approach for data preparation and model validation. Surg Endosc 36, 3811–3821 (2022). https://doi.org/10.1007/s00464-021-08698-2
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00464-021-08698-2