Application of artificial intelligence for endoscopic image analysis in inflammatory bowel diseases


如何引用文章

全文:

开放存取 开放存取
受限制的访问 ##reader.subscriptionAccessGranted##
受限制的访问 订阅或者付费存取

详细

Ulcerative colitis (UC) and Crohn's disease (CD) often lead to the development of complications when the basic therapy is delayed. The delay in disease verification is up to 2 years for CD and up to 10 months for UC. Up to 25% of cases of inflammatory bowel disease (IBD) are diagnosed 2 years after the onset of symptoms. In recent years, methods for detecting IBD based on artificial networks have been created. They increase the accuracy of diagnosis and thereby improve the prognosis of patients. The aim of the study: to create a method for the diagnosis and differential diagnosis of IBD based on the analysis of the artificial networks of endoscopic images. Material and methods. The study included patients aged 18 years with colonic CD and UC after excluding intestinal infections, who had endoscopic exacerbation. The comparison group consisted of patients with visually unchanged colon mucosa. Patients underwent videocolonoscopy, during which the obtained digital images were filtered using a non-local mean filter and contrast improved by adaptive contrast-limited histogram equalization. We developed the artificial networks based on the VGG16 network: the first network determined the presence of mucosal changes, the second gave a conclusion about the type of IBD. The input element was digital images, and the output layer of the artificial network gave a conclusion about the presence of pathology and the type of IBD. Results. The artificial network detected pathology with an accuracy of 89,3% and differentiated IBD with an accuracy of 81,9%. The artificial network detected the norm with an accuracy of 88% and, in comparison with other classes, determined the UC better (90% accuracy), and also most well distinguished the CD class from other classes (92% accuracy). The overall accuracy was 84,6%. On the basis of the developed artificial network, the doctor's decision support system (DDSS) was created. Conclusion. The developed models showed moderate accuracy in the detection of IBD and high accuracy in the determination of UC and CD. The created DDSS can reduce the time of IBD verification and can be used as an additional tool in the practice of physicians.

全文:

受限制的访问

作者简介

Igor Bakulin

I.I. Mechnikov North-Western State Medical University of the Ministry of Healthcare of Russia

Email: igbakulin@yandex.ru
Dr. med. habil., professor, head of the Department of propaedeutics of internal diseases, gastroenterology and dietology named after S.M. Ryss

Irina Rasmagina

I.I. Mechnikov North-Western State Medical University of the Ministry of Healthcare of Russia

Email: irenerasmagina@gmail.com
postgraduate student of the 2nd year of study in the specialty «Internal medicine»

Maria Skalinskaya

I.I. Mechnikov North-Western State Medical University of the Ministry of Healthcare of Russia

Email: mskalinskaya@yahoo.com
PhD in Medicine, associate professor, associate professor of the Department of propaedeutics of internal diseases, gastroenterology and dietology named after S.M. Ryss

Gleb Mashevskiy

V.I. Ulyanov (Lenin) Saint Petersburg State Electrotechnical University «LETI»

Email: aniket@list.ru
PhD in Medicine, associate professor of the Department of biotechnical systems 197022, Saint Petersburg, 5 Professora Popova Str

Natalia Shelyakina

Email: n.sheliakina@gmail.com
systems analyst.

参考

  1. Клинические рекомендации. Язвенный колит. Российская гастроэнтерологическая ассоциация, общероссийская общественная организация «Ассоциация колопроктологов России». Рубрикатор клинических рекомендаций Минздрава России. 2020. ID: 193. Доступ: https://cr.minzdrav.gov.ru/schema/193_1 (дата обращения - 01.08.2022).
  2. Клинические рекомендации. Болезнь Крона. Российская гастроэнтерологическая ассоциация, общероссийская общественная организация «Ассоциация колопроктологов России». Рубрикатор клинических рекомендаций Минздрава России. 2020. ID: 176. Доступ: https://cr.minzdrav.gov.ru/schema/176_1 (дата обращения - 01.08.2022).
  3. Novacek G., Grochenig H.P., Haas T. et al. Diagnostic delay in patients with inflammatory bowel disease in Austria. Wien Klin Wochenschr. 2019; 131(5-6): 104-12. https://dx.doi.org/10.1007/s00508-019-1451-3.
  4. Cantoro L., Di Sabatino A., Papi C. et al. The time course of diagnostic delay in inflammatory bowel disease over the last sixty years: An Italian multicentre study. J Crohns Colitis. 2017; 11(8): 975-80. https://dx.doi.org/10.1093/ecco-jcc/jjx041.
  5. Walker G.J., Lin S., Chanchlani N. et al. Quality improvement project identifies factors associated with delay in IBD diagnosis. Aliment Pharmacol Ther. 2020; 52(3): 471-80. https://dx.doi.org/10.1111/apt.15885.
  6. Vavricka S.R., Spigaglia S.M., Rogler G. et al. Systematic evaluation of risk factors for diagnostic delay in inflammatory bowel disease. Inflamm Bowel Dis. 2012; 18(3): 496-505. https://dx.doi.org/10.1002/ibd.21719.
  7. Zaharie R., Tantau A., Zaharie F. et al. Diagnostic delay in Romanian patients with inflammatory bowel disease: Risk factors and impact on the disease course and need for surgery. J Crohns Colitis. 2016; 10(3): 306-14. https://dx.doi.org/10.1093/ecco-jcc/jjv215.
  8. Lee D.W., Koo J.S., Choe J.W. et al. Diagnostic delay in inflammatory bowel disease increases the risk of intestinal surgery. World J Gastroenterol. 2017; 23(35): 6474-81. https://dx.doi.org/10.3748/wjg.v23.i35.6474.
  9. Schoepfer A.M., Dehlavi M.A., Fournier N. et al. Diagnostic delay in Crohn's disease is associated with a complicated disease course and increased operation rate. Am J Gastroenterol. 2013; 108(11): 1744-53; quiz 1754. https://dx.doi.org/10.1038/ajg.2013.248.
  10. Маев И.В., Шелыгин Ю.А., Скалинская М.И. с соавт. Патоморфоз воспалительных заболеваний кишечника. Вестник Российской академии медицинских наук. 2020; 75(1): 27-35.
  11. Kroner P.T., Engels M.M., Glicksberg B.S. et al. Artificial intelligence in gastroenterology: A state-of-the-art review. World J Gastroenterol. 2021; 27(40): 6794-824. https://dx.doi.org/10.3748/wjg.v27.i40.6794.
  12. Gubatan J., Levitte S., Patel A. et al. Artificial intelligence applications in inflammatory bowel disease: Emerging technologies and future directions. World J Gastroenterol. 2021; 27(17): 1920-35. https://dx.doi.org/10.3748/wjg.v27.i17.1920.
  13. Abadir A.P., Ali M.F., Karnes W., Samarasena J.B. Artificial intelligence in gastrointestinal endoscopy. Clin Endosc. 2020; 53(2): 132-41. https://dx.doi.org/10.5946/ce.2020.038.
  14. Sturm A., Maaser C., Calabrese E. et al. ECCO-ESGAR guideline for diagnostic assessment in IBD Part 2: IBD scores and general principles and technical aspects. J Crohns Colitis. 2019; 13(3): 273-84. https://dx.doi.org/10.1093/ecco-jcc/jjy114.
  15. Buades A., Coll B., Morel J.M. A Non-local algorithm for image denoising. In: 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05). Vol. 2. Washington; IEEE Computer Society. 2005: 60-65. https://dx.doi.org/10.1109/CVPR.2005.38. Print ISSN: 1063-6919.
  16. Zuiderveld K. Contrast limited adaptive histogram equalization. In: Graphics Gems IV. San Diego: Academic Press Professional, Inc. 1994; 474-85. ISBN: 978-0-12-336155-4.
  17. Simonyan K., Zisserman A. Very deep convolutional networks for large-scale image recognition. Published online April 10, 2015. URL: http://arxiv.org/abs/1409.1556 (date of access - 01.08.2022).
  18. Бакулин И.Г., Расмагина И.А., Скалинская М.И. Дифференциальная диагностика и прогнозирование течения воспалительных заболеваний кишечника: современные подходы. Вестник Северо-Западного государственного медицинского университета им. И.И. Мечникова. 2021; 13(3): 19-30. https://dx.doi.org/10.17816/mechnikov77646. EDN: YEILRQ.
  19. Ruan G., Qi J., Cheng Y. et al. Development and validation of a deep neural network for accurate identification of endoscopic images from patients with ulcerative colitis and Crohn's disease. Front Med (Lausanne). 2022; 9: 854677. https://dx.doi.org/10.3389/fmed.2022.854677.
  20. Tong Y., Lu K., Yang Y. et al. Can natural language processing help differentiate inflammatory intestinal diseases in China? Models applying random forest and convolutional neural network approaches. BMC Med Inform Decis Mak. 2020; 20(1): 248. https://dx.doi.org/10.1186/s12911-020-01277-w.
  21. Aoki T., Yamada A., Aoyama K. et al. Automatic detection of erosions and ulcerations in wireless capsule endoscopy images based on a deep convolutional neural network. Gastrointest Endosc. 2019; 89(2): 357-63.e2. https://dx.doi.org/10.1016/j.gie.2018.10.027.

补充文件

附件文件
动作
1. JATS XML
下载

版权所有 © Bionika Media, 2022
##common.cookie##