Study of Algorithms Based on YOLO Neural Network Architecture in the Problem of Polyp Detection on Colonoscopic Video Data-Reference-Cited by-同舟云学术

Study of Algorithms Based on YOLO Neural Network Architecture in the Problem of Polyp Detection on Colonoscopic Video Data

Published:2023 Issue: Volume: Page:
ISSN:
Container-title:Proceedings of the 33rd International Conference on Computer Graphics and Vision
language:
Short-container-title:

Author:

Khryashev V.V.^ORCID,Kotov N.V.^ORCID,Priorov A.L.^ORCID

Abstract

An analysis was made of the use of neural network algorithms for the detection of colon polyps obtained during colonoscopy. The Kvasir-SEG image database was used to train and test deep machine learning algorithms. The networks YOLOv6, YOLOR, YOLOv7, YOLOv7X, YOLOv8 previously trained on the basis of MS COCO images were used as neural network architectures. Due to the small volume of images in the Kvasir-SEG database, data augmentation was used. As a result of applying the detection algorithms to the test set of endoscopic images, the highest values of the metrics AP@[0,25..0,75] equal to 98,4 and AP@0,50 equal to 98,6 were obtained for the neural network detector based on the YOLOv8 network. According to the results of comparing the proposed algorithms with analogues, the YOLOv8 assessment showed an increase in the results for the AP@[0.25..0.75] metric by 5.9 in searches with the previous model НАОО<sup>YL</sup>. The results obtained can be used in the development of a video stream analysis system in an endoscopic system operating in real time during colonoscopy studies

Publisher

Keldysh Institute of Applied Mathematics

Reference22 articles.

1. Палевская С.А., Короткевич А.Г. Эндоскопия желудочно-кишечного тракта. М.: ГЭОТАР-Медиа, 2020. 752 c.

2. Куваев Р.О., Кашин С.В. Современное эндоскопическое исследование желудка с использованием методик узкоспектральной и увеличительной эндоскопии: техника проведения и алгоритмы диагностики // Доказательная гастроэнтерология. 2016. Т. 2, № 5. С. 3–13.

3. Lee J. Resection of diminutive and small colorectal polyps: what is the optimal technique? Clinical endoscopy. 2016. vol. 49(4). pp. 355.

4. Convolutional neural network for early detection of gastric cancer by endoscopic video analysis / A. Lebedev [et al.] // Proc. SPIE 11433. Twelfth International Conference on Machine Vision (ICMV 2019).

5. Батухтин Д.М., Пеганова Е.В., Митракова Н.Н. Анализ узкоспектральных эндоскопических изображений на внутренней поверхности пищевода // Вестник Поволжского государственного технологического университета. Серия: радиотехнические и инфокоммуникационные системы. 2014. № 4 (23). С. 45–57.