No Cover Image

Journal article 365 views 23 downloads

An End-to-End Real-Time Lightweight Network for the Joint Segmentation of Optic Disc and Optic Cup on Fundus Images

Zhijie Liu, Yuanqiong Chen, Xiaohua Xiang, Zhan Li, Bolin Liao, Jianfeng Li

Mathematics, Volume: 10, Issue: 22, Start page: 4288

Swansea University Author: Zhan Li

  • 62097.pdf

    PDF | Version of Record

    © 2022 by the authors. This is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license

    Download (3.12MB)

Check full text

DOI (Published version): 10.3390/math10224288

Abstract

Glaucoma is the second-most-blinding eye disease in the world and accurate segmentation of the optic disc (OD) and optic cup (OC) is essential for the diagnosis of glaucoma. To solve the problems of poor real-time performance, high algorithm complexity, and large memory consumption of fundus segment...

Full description

Published in: Mathematics
ISSN: 2227-7390
Published: MDPI AG 2022
Online Access: Check full text

URI: https://cronfa.swan.ac.uk/Record/cronfa62097
Tags: Add Tag
No Tags, Be the first to tag this record!
first_indexed 2022-12-01T09:27:05Z
last_indexed 2023-01-13T19:23:19Z
id cronfa62097
recordtype SURis
fullrecord <?xml version="1.0"?><rfc1807><datestamp>2023-01-10T09:01:21.1461334</datestamp><bib-version>v2</bib-version><id>62097</id><entry>2022-12-01</entry><title>An End-to-End Real-Time Lightweight Network for the Joint Segmentation of Optic Disc and Optic Cup on Fundus Images</title><swanseaauthors><author><sid>94f19a09e17bad497ef1b4a0992c1d56</sid><firstname>Zhan</firstname><surname>Li</surname><name>Zhan Li</name><active>true</active><ethesisStudent>false</ethesisStudent></author></swanseaauthors><date>2022-12-01</date><deptcode>SCS</deptcode><abstract>Glaucoma is the second-most-blinding eye disease in the world and accurate segmentation of the optic disc (OD) and optic cup (OC) is essential for the diagnosis of glaucoma. To solve the problems of poor real-time performance, high algorithm complexity, and large memory consumption of fundus segmentation algorithms, a lightweight segmentation algorithm, GlauNet, based on convolutional neural networks, is proposed. The algorithm designs an efficient feature-extraction network and proposes a multiscale boundary fusion (MBF) module, which greatly improves the segmentation efficiency of the algorithm while ensuring segmentation accuracy. Experiments show that the algorithm achieves Dice scores of 0.9701/0.8959, 0.9650/0.8621, and 0.9594/0.8795 on three publicly available datasets&#x2014;Drishti-GS, RIM-ONE-r3, and REFUGE-train&#x2014;for both the optic disc and the optic cup. The number of model parameters is only 0.8 M, and it only takes 13 ms to infer an 800 &#xD7; 800 fundus image on a GTX 3070 GPU.</abstract><type>Journal Article</type><journal>Mathematics</journal><volume>10</volume><journalNumber>22</journalNumber><paginationStart>4288</paginationStart><paginationEnd/><publisher>MDPI AG</publisher><placeOfPublication/><isbnPrint/><isbnElectronic/><issnPrint/><issnElectronic>2227-7390</issnElectronic><keywords>convolutional neural network; optic disc and cup segmentation; glaucoma screening; medical auxiliary diagnosis</keywords><publishedDay>16</publishedDay><publishedMonth>11</publishedMonth><publishedYear>2022</publishedYear><publishedDate>2022-11-16</publishedDate><doi>10.3390/math10224288</doi><url/><notes/><college>COLLEGE NANME</college><department>Computer Science</department><CollegeCode>COLLEGE CODE</CollegeCode><DepartmentCode>SCS</DepartmentCode><institution>Swansea University</institution><apcterm/><funders>The work was supported by National Natural Science Foundation of China (61962023).</funders><projectreference/><lastEdited>2023-01-10T09:01:21.1461334</lastEdited><Created>2022-12-01T09:23:37.5049039</Created><path><level id="1">Faculty of Science and Engineering</level><level id="2">School of Mathematics and Computer Science - Computer Science</level></path><authors><author><firstname>Zhijie</firstname><surname>Liu</surname><order>1</order></author><author><firstname>Yuanqiong</firstname><surname>Chen</surname><order>2</order></author><author><firstname>Xiaohua</firstname><surname>Xiang</surname><order>3</order></author><author><firstname>Zhan</firstname><surname>Li</surname><order>4</order></author><author><firstname>Bolin</firstname><surname>Liao</surname><order>5</order></author><author><firstname>Jianfeng</firstname><surname>Li</surname><order>6</order></author></authors><documents><document><filename>62097__25965__599d5c197491403d9cac0b99c3e01804.pdf</filename><originalFilename>62097.pdf</originalFilename><uploaded>2022-12-01T09:26:16.5244926</uploaded><type>Output</type><contentLength>3272751</contentLength><contentType>application/pdf</contentType><version>Version of Record</version><cronfaStatus>true</cronfaStatus><documentNotes>&#xA9; 2022 by the authors. This is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license</documentNotes><copyrightCorrect>true</copyrightCorrect><language>eng</language><licence>https://creativecommons.org/licenses/by/4.0/</licence></document></documents><OutputDurs/></rfc1807>
spelling 2023-01-10T09:01:21.1461334 v2 62097 2022-12-01 An End-to-End Real-Time Lightweight Network for the Joint Segmentation of Optic Disc and Optic Cup on Fundus Images 94f19a09e17bad497ef1b4a0992c1d56 Zhan Li Zhan Li true false 2022-12-01 SCS Glaucoma is the second-most-blinding eye disease in the world and accurate segmentation of the optic disc (OD) and optic cup (OC) is essential for the diagnosis of glaucoma. To solve the problems of poor real-time performance, high algorithm complexity, and large memory consumption of fundus segmentation algorithms, a lightweight segmentation algorithm, GlauNet, based on convolutional neural networks, is proposed. The algorithm designs an efficient feature-extraction network and proposes a multiscale boundary fusion (MBF) module, which greatly improves the segmentation efficiency of the algorithm while ensuring segmentation accuracy. Experiments show that the algorithm achieves Dice scores of 0.9701/0.8959, 0.9650/0.8621, and 0.9594/0.8795 on three publicly available datasets—Drishti-GS, RIM-ONE-r3, and REFUGE-train—for both the optic disc and the optic cup. The number of model parameters is only 0.8 M, and it only takes 13 ms to infer an 800 × 800 fundus image on a GTX 3070 GPU. Journal Article Mathematics 10 22 4288 MDPI AG 2227-7390 convolutional neural network; optic disc and cup segmentation; glaucoma screening; medical auxiliary diagnosis 16 11 2022 2022-11-16 10.3390/math10224288 COLLEGE NANME Computer Science COLLEGE CODE SCS Swansea University The work was supported by National Natural Science Foundation of China (61962023). 2023-01-10T09:01:21.1461334 2022-12-01T09:23:37.5049039 Faculty of Science and Engineering School of Mathematics and Computer Science - Computer Science Zhijie Liu 1 Yuanqiong Chen 2 Xiaohua Xiang 3 Zhan Li 4 Bolin Liao 5 Jianfeng Li 6 62097__25965__599d5c197491403d9cac0b99c3e01804.pdf 62097.pdf 2022-12-01T09:26:16.5244926 Output 3272751 application/pdf Version of Record true © 2022 by the authors. This is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license true eng https://creativecommons.org/licenses/by/4.0/
title An End-to-End Real-Time Lightweight Network for the Joint Segmentation of Optic Disc and Optic Cup on Fundus Images
spellingShingle An End-to-End Real-Time Lightweight Network for the Joint Segmentation of Optic Disc and Optic Cup on Fundus Images
Zhan Li
title_short An End-to-End Real-Time Lightweight Network for the Joint Segmentation of Optic Disc and Optic Cup on Fundus Images
title_full An End-to-End Real-Time Lightweight Network for the Joint Segmentation of Optic Disc and Optic Cup on Fundus Images
title_fullStr An End-to-End Real-Time Lightweight Network for the Joint Segmentation of Optic Disc and Optic Cup on Fundus Images
title_full_unstemmed An End-to-End Real-Time Lightweight Network for the Joint Segmentation of Optic Disc and Optic Cup on Fundus Images
title_sort An End-to-End Real-Time Lightweight Network for the Joint Segmentation of Optic Disc and Optic Cup on Fundus Images
author_id_str_mv 94f19a09e17bad497ef1b4a0992c1d56
author_id_fullname_str_mv 94f19a09e17bad497ef1b4a0992c1d56_***_Zhan Li
author Zhan Li
author2 Zhijie Liu
Yuanqiong Chen
Xiaohua Xiang
Zhan Li
Bolin Liao
Jianfeng Li
format Journal article
container_title Mathematics
container_volume 10
container_issue 22
container_start_page 4288
publishDate 2022
institution Swansea University
issn 2227-7390
doi_str_mv 10.3390/math10224288
publisher MDPI AG
college_str Faculty of Science and Engineering
hierarchytype
hierarchy_top_id facultyofscienceandengineering
hierarchy_top_title Faculty of Science and Engineering
hierarchy_parent_id facultyofscienceandengineering
hierarchy_parent_title Faculty of Science and Engineering
department_str School of Mathematics and Computer Science - Computer Science{{{_:::_}}}Faculty of Science and Engineering{{{_:::_}}}School of Mathematics and Computer Science - Computer Science
document_store_str 1
active_str 0
description Glaucoma is the second-most-blinding eye disease in the world and accurate segmentation of the optic disc (OD) and optic cup (OC) is essential for the diagnosis of glaucoma. To solve the problems of poor real-time performance, high algorithm complexity, and large memory consumption of fundus segmentation algorithms, a lightweight segmentation algorithm, GlauNet, based on convolutional neural networks, is proposed. The algorithm designs an efficient feature-extraction network and proposes a multiscale boundary fusion (MBF) module, which greatly improves the segmentation efficiency of the algorithm while ensuring segmentation accuracy. Experiments show that the algorithm achieves Dice scores of 0.9701/0.8959, 0.9650/0.8621, and 0.9594/0.8795 on three publicly available datasets—Drishti-GS, RIM-ONE-r3, and REFUGE-train—for both the optic disc and the optic cup. The number of model parameters is only 0.8 M, and it only takes 13 ms to infer an 800 × 800 fundus image on a GTX 3070 GPU.
published_date 2022-11-16T04:21:27Z
_version_ 1763754416299573248
score 10.997593

Similar Items