SIA OpenIR  > 机器人学研究室
Multi-Scale Context-Guided Deep Network for Automated Lesion Segmentation with Endoscopy Images of Gastrointestinal Tract
Wang S(王帅)1; Cong Y(丛杨)1; Zhu, Hancan2; Chen, Xianyi3; Qu LQ(屈靓琼)1; Fan HJ(范慧杰)1; Zhang, Qiang4; Liu, Mingxia5
Department机器人学研究室
Source PublicationIEEE Journal of Biomedical and Health Informatics
ISSN2168-2194
2021
Volume25Issue:2Pages:514-525
Indexed BySCI ; EI
EI Accession number20210709918704
WOS IDWOS:000616310200021
Contribution Rank1
Funding OrganizationNational Natural Science Foundation of China under Grants 61703301 and 61602307 ; Natural Science Foundation of Zhejiang Province under Grant LY19F020013 ; Taishan Scholar Program of Shandong Province in China ; Shandong Provincial Natural Science Foundation under Grant ZR2019YQ27 ; Scientific Research Foundation of Taishan University under Grant Y-01-2018019
KeywordMulti-scale Context fully convolutional network lesion segmentation endoscopy image gastrointestinal tract
Abstract

Accurate lesion segmentation based on endoscopy images is a fundamental task for the automated diagnosis of gastrointestinal tract (GI Tract) diseases. Previous studies usually use hand-crafted features for representing endoscopy images, while feature definition and lesion segmentation are treated as two standalone tasks. Due to the possible heterogeneity between features and segmentation models, these methods often result in sub-optimal performance. Several fully convolutional networks have been recently developed to jointly perform feature learning and model training for GI Tract disease diagnosis. However, they generally ignore local spatial details of endoscopy images, as down-sampling operations (e.g., pooling and convolutional striding) may result in irreversible loss of image spatial information. To this end, we propose a multi-scale context-guided deep network (MCNet) for end-to-end lesion segmentation of endoscopy images in GI Tract, where both global and local contexts are captured as guidance for model training. Specifically, one global subnetwork is designed to extract the global structure and high-level semantic context of each input image. Then we further design two cascaded local subnetworks based on output feature maps of the global subnetwork, aiming to capture both local appearance information and relatively high-level semantic information in a multi-scale manner. Those feature maps learned by three subnetworks are further fused for the subsequent task of lesion segmentation. We have evaluated the proposed MCNet on 1,310 endoscopy images from the public EndoVis-Ab and CVC-ClinicDB datasets for abnormal segmentation and polyp segmentation, respectively. Experimental results demonstrate that MCNet achieves text{74}% and text{85}% mean intersection over union (mIoU) on two datasets, respectively, outperforming several state-of-the-art approaches in automated lesion segmentation with endoscopy images of GI Tract.

Language英语
WOS SubjectComputer Science, Information Systems ; Computer Science, Interdisciplinary Applications ; Mathematical & Computational Biology ; Medical Informatics
WOS Research AreaComputer Science ; Mathematical & Computational Biology ; Medical Informatics
Funding ProjectNational Natural Science Foundation of China[61703301] ; National Natural Science Foundation of China[61602307] ; Natural Science Foundation of Zhejiang Province[LY19F020013] ; Taishan Scholar Program of Shandong Province in China ; Shandong Provincial Natural Science Foundation[ZR2019YQ27] ; Scientific Research Foundation of Taishan University[Y-01-2018019]
Citation statistics
Document Type期刊论文
Identifierhttp://ir.sia.cn/handle/173321/28339
Collection机器人学研究室
Corresponding AuthorLiu, Mingxia
Affiliation1.State Key Laboratory of Robotics, Shenyang Institute of Automation, Chinese Academy of Sciences, Shenyang, China
2.School of Mathematics Physics and Information, Shaoxing University, Shaoxing, China
3.School of Computer and Software, Nanjing University of Information Science and Technology, Nanjing, China
4.School of Computer Science and Technology, Dalian University of Technology, Dalian, China
5.Department of Information Science and Technology, Taishan Univeristy, Taian, China
Recommended Citation
GB/T 7714
Wang S,Cong Y,Zhu, Hancan,et al. Multi-Scale Context-Guided Deep Network for Automated Lesion Segmentation with Endoscopy Images of Gastrointestinal Tract[J]. IEEE Journal of Biomedical and Health Informatics,2021,25(2):514-525.
APA Wang S.,Cong Y.,Zhu, Hancan.,Chen, Xianyi.,Qu LQ.,...&Liu, Mingxia.(2021).Multi-Scale Context-Guided Deep Network for Automated Lesion Segmentation with Endoscopy Images of Gastrointestinal Tract.IEEE Journal of Biomedical and Health Informatics,25(2),514-525.
MLA Wang S,et al."Multi-Scale Context-Guided Deep Network for Automated Lesion Segmentation with Endoscopy Images of Gastrointestinal Tract".IEEE Journal of Biomedical and Health Informatics 25.2(2021):514-525.
Files in This Item:
File Name/Size DocType Version Access License
Multi-Scale Context-(4432KB)期刊论文出版稿开放获取CC BY-NC-SAView Application Full Text
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[Wang S(王帅)]'s Articles
[Cong Y(丛杨)]'s Articles
[Zhu, Hancan]'s Articles
Baidu academic
Similar articles in Baidu academic
[Wang S(王帅)]'s Articles
[Cong Y(丛杨)]'s Articles
[Zhu, Hancan]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[Wang S(王帅)]'s Articles
[Cong Y(丛杨)]'s Articles
[Zhu, Hancan]'s Articles
Terms of Use
No data!
Social Bookmark/Share
File name: Multi-Scale Context-Guided Deep Network for Automated Lesion Segmentation with Endoscopy Images of Gastrointestinal Tract.pdf
Format: Adobe PDF
All comments (0)
No comment.
 

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.