Automatic detection of different types of small-bowel lesions on capsule endoscopy images using a newly developed deep convolutional neural network

Keita Otani; Ayako Nakada; Yusuke Kurose; Ryota Niikura; Atsuo Yamada; Tomonori Aoki; Hiroyoshi Nakanishi; Hisashi Doyama; Kenkei Hasatani; Tetsuya Sumiyoshi; Masaru Kitsuregawa; Tatsuya Harada; Kazuhiko Koike

doi:10.1055/a-1167-8157

Endoscopy, Table of Contents

Endoscopy 2020; 52(09): 786-791
DOI: 10.1055/a-1167-8157

Innovations and brief communications

Automatic detection of different types of small-bowel lesions on capsule endoscopy images using a newly developed deep convolutional neural network

Keita Otani

¹Department of Mechano-Informatics, Graduate School of Information Science and Technology, The University of Tokyo, Tokyo, Japan

,

Ayako Nakada

²Department of Gastroenterology, Graduate School of Medicine, The University of Tokyo, Tokyo, Japan

,

Yusuke Kurose

¹Department of Mechano-Informatics, Graduate School of Information Science and Technology, The University of Tokyo, Tokyo, Japan

³Center for Advanced Intelligence Project, RIKEN, Tokyo, Japan

,

Ryota Niikura

²Department of Gastroenterology, Graduate School of Medicine, The University of Tokyo, Tokyo, Japan

,

Atsuo Yamada

²Department of Gastroenterology, Graduate School of Medicine, The University of Tokyo, Tokyo, Japan

,

Tomonori Aoki

²Department of Gastroenterology, Graduate School of Medicine, The University of Tokyo, Tokyo, Japan

,

Hiroyoshi Nakanishi

⁴Department of Gastroenterology, Ishikawa Prefectural Central Hospital, Kanazawa-shi, Ishikawa, Japan

,

Hisashi Doyama

⁴Department of Gastroenterology, Ishikawa Prefectural Central Hospital, Kanazawa-shi, Ishikawa, Japan

,

Kenkei Hasatani

⁵Department of Gastroenterology, Fukui Prefectural Hospital, Fukui-shi, Fukui, Japan

,

Tetsuya Sumiyoshi

⁶The Center for Digestive Disease, Tonan Hospital, Sapporo-shi, Hokkaido, Japan

,

Masaru Kitsuregawa

⁷Institute of Industrial Science, The University of Tokyo, Tokyo, Japan

⁸National Institute of Informatics, Tokyo, Japan

,

Tatsuya Harada

³Center for Advanced Intelligence Project, RIKEN, Tokyo, Japan

⁹Research Center for Advanced Science and Technology, The University of Tokyo, Tokyo, Japan

¹⁰Research Center for Medical Bigdata, National Institute of Informatics, Tokyo, Japan

,

Kazuhiko Koike

²Department of Gastroenterology, Graduate School of Medicine, The University of Tokyo, Tokyo, Japan

› Author Affiliations

Abstract

Background Previous computer-aided detection systems for diagnosing lesions in images from wireless capsule endoscopy (WCE) have been limited to a single type of small-bowel lesion. We developed a new artificial intelligence (AI) system able to diagnose multiple types of lesions, including erosions and ulcers, vascular lesions, and tumors.

Methods We trained the deep neural network system RetinaNet on a data set of 167 patients, which consisted of images of 398 erosions and ulcers, 538 vascular lesions, 4590 tumors, and 34 437 normal tissues. We calculated the mean area under the receiver operating characteristic curve (AUC) for each lesion type using five-fold stratified cross-validation.

Results The mean age of the patients was 63.6 years; 92 were men. The mean AUCs of the AI system were 0.996 (95 %CI 0.992 – 0.999) for erosions and ulcers, 0.950 (95 %CI 0.923 – 0.978) for vascular lesions, and 0.950 (95 %CI 0.913 – 0.988) for tumors.

Conclusion We developed and validated a new computer-aided diagnosis system for multiclass diagnosis of small-bowel lesions in WCE images.

Full Text

References

References
1 Iddan G, Meron G, Glukhovsky A. et al. Wireless capsule endoscopy. Nature 2000; 405: 417
2 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: 357-363
3 Tsuboi A, Oka S, Aoyama K. et al. Artificial intelligence using a convolutional neural network for automatic detection of small-bowel angioectasia in capsule endoscopy images. Dig Endosc 2020; 32: 382-390
4 Zhou T, Han G, Li BN. et al. Quantitative analysis of patients with celiac disease by video capsule endoscopy: a deep learning method. Comput Biol Med 2017; 85: 1-6
5 Ding Z, Shi H, Zhang H. et al. Gastroenterologist-level identification of small-bowel diseases and normal variants by capsule endoscopy using a deep-learning model. Gastroenterology 2019; 157: 1044-1054
6 Albert J, Gobel CM, Lesske J. et al. Simethicone for small bowel preparation for capsule endoscopy: a systematic, single-blinded, controlled study. Gastrointest Endosc 2004; 59: 487-491
7 Niikura R, Yamada A, Maki K. et al. Associations between drugs and small-bowel mucosal bleeding: multicenter capsule-endoscopy study. Dig Endosc 2018; 30: 79-89
8 Liu W, Anguelov D, Erhan D. et al. SSD: Single Shot MultiBox Detector. In: Leibe B, Matas J, Sebe N, Welling M. eds. Computer Vision – ECCV 2016. Proceedings of the 14th European Conference on Computer Vision; 2016 Oct 11–14. Amsterdam, The Netherlands:
9 Lin T, Goyal P, Girshick R. et al. Focal loss for dense object detection. In: Fangyu L, Shuaipeng L, Liqiang Z. et al. eds. IEEE International Conference on Computer Vision (ICCV) 2017. Proceedings of the 16th ICCV; 2017 Oct 22–29. Venice, Italy: 2980-2988
10 DeLong ER, DeLong DM, Clarke-Pearson DL. Comparing the areas under two or more correlated receiver operating characteristic curves: a nonparametric approach. Biometrics 1988; 44: 837-845

Supplementary Material

Supplementary material