MSM-TDE: multi-scale semantics mining and tiny details enhancement network for retinal vessel segmentation
Abstract Retinal image segmentation is crucial for the early diagnosis of some diseases like diabetes and hypertension. Current methods face many challenges, such as inadequate multi-scale semantics and insufficient global information. In view of this, we propose a network called multi-scale semanti...
Saved in:
Main Authors: | , , , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Springer
2025-01-01
|
Series: | Complex & Intelligent Systems |
Subjects: | |
Online Access: | https://doi.org/10.1007/s40747-024-01714-7 |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
_version_ | 1832571184743448576 |
---|---|
author | Hongbin Zhang Jin Zhang Xuan Zhong Ya Feng Guangli Li Xiong Li Jingqin Lv Donghong Ji |
author_facet | Hongbin Zhang Jin Zhang Xuan Zhong Ya Feng Guangli Li Xiong Li Jingqin Lv Donghong Ji |
author_sort | Hongbin Zhang |
collection | DOAJ |
description | Abstract Retinal image segmentation is crucial for the early diagnosis of some diseases like diabetes and hypertension. Current methods face many challenges, such as inadequate multi-scale semantics and insufficient global information. In view of this, we propose a network called multi-scale semantics mining and tiny details enhancement (MSM-TDE). First, a multi-scale feature input module is designed to capture multi-scale semantics information from the source. Then a fresh multi-scale attention guidance module is constructed to mine local multi-scale semantics while a global semantics enhancement module is proposed to extract global multi-scale semantics. Additionally, an auxiliary vessel detail enhancement branch using dynamic snake convolution is built to enhance the tiny vessel details. Extensive experimental results on four public datasets validate the superiority of MSM-TDE, which obtains competitive performance with satisfactory model complexity. Notably, this study provides an innovative idea of multi-scale semantics mining by diverse methods. |
format | Article |
id | doaj-art-97ecee0f8502459dbdacb75e2c40f0bb |
institution | Kabale University |
issn | 2199-4536 2198-6053 |
language | English |
publishDate | 2025-01-01 |
publisher | Springer |
record_format | Article |
series | Complex & Intelligent Systems |
spelling | doaj-art-97ecee0f8502459dbdacb75e2c40f0bb2025-02-02T12:49:22ZengSpringerComplex & Intelligent Systems2199-45362198-60532025-01-0111112310.1007/s40747-024-01714-7MSM-TDE: multi-scale semantics mining and tiny details enhancement network for retinal vessel segmentationHongbin Zhang0Jin Zhang1Xuan Zhong2Ya Feng3Guangli Li4Xiong Li5Jingqin Lv6Donghong Ji7School of Information and Software Engineering, East China Jiaotong UniversitySchool of Information and Software Engineering, East China Jiaotong UniversityModern Industry School of Virtual Reality, Jiangxi University of Finance and EconomicsSchool of Information and Software Engineering, East China Jiaotong UniversitySchool of Information and Software Engineering, East China Jiaotong UniversitySchool of Information and Software Engineering, East China Jiaotong UniversitySchool of Information and Software Engineering, East China Jiaotong UniversityCyber Science and Engineering School, Wuhan UniversityAbstract Retinal image segmentation is crucial for the early diagnosis of some diseases like diabetes and hypertension. Current methods face many challenges, such as inadequate multi-scale semantics and insufficient global information. In view of this, we propose a network called multi-scale semantics mining and tiny details enhancement (MSM-TDE). First, a multi-scale feature input module is designed to capture multi-scale semantics information from the source. Then a fresh multi-scale attention guidance module is constructed to mine local multi-scale semantics while a global semantics enhancement module is proposed to extract global multi-scale semantics. Additionally, an auxiliary vessel detail enhancement branch using dynamic snake convolution is built to enhance the tiny vessel details. Extensive experimental results on four public datasets validate the superiority of MSM-TDE, which obtains competitive performance with satisfactory model complexity. Notably, this study provides an innovative idea of multi-scale semantics mining by diverse methods.https://doi.org/10.1007/s40747-024-01714-7Retinal vessel segmentationMulti-scale semantics miningTiny details enhancementU-Net |
spellingShingle | Hongbin Zhang Jin Zhang Xuan Zhong Ya Feng Guangli Li Xiong Li Jingqin Lv Donghong Ji MSM-TDE: multi-scale semantics mining and tiny details enhancement network for retinal vessel segmentation Complex & Intelligent Systems Retinal vessel segmentation Multi-scale semantics mining Tiny details enhancement U-Net |
title | MSM-TDE: multi-scale semantics mining and tiny details enhancement network for retinal vessel segmentation |
title_full | MSM-TDE: multi-scale semantics mining and tiny details enhancement network for retinal vessel segmentation |
title_fullStr | MSM-TDE: multi-scale semantics mining and tiny details enhancement network for retinal vessel segmentation |
title_full_unstemmed | MSM-TDE: multi-scale semantics mining and tiny details enhancement network for retinal vessel segmentation |
title_short | MSM-TDE: multi-scale semantics mining and tiny details enhancement network for retinal vessel segmentation |
title_sort | msm tde multi scale semantics mining and tiny details enhancement network for retinal vessel segmentation |
topic | Retinal vessel segmentation Multi-scale semantics mining Tiny details enhancement U-Net |
url | https://doi.org/10.1007/s40747-024-01714-7 |
work_keys_str_mv | AT hongbinzhang msmtdemultiscalesemanticsminingandtinydetailsenhancementnetworkforretinalvesselsegmentation AT jinzhang msmtdemultiscalesemanticsminingandtinydetailsenhancementnetworkforretinalvesselsegmentation AT xuanzhong msmtdemultiscalesemanticsminingandtinydetailsenhancementnetworkforretinalvesselsegmentation AT yafeng msmtdemultiscalesemanticsminingandtinydetailsenhancementnetworkforretinalvesselsegmentation AT guanglili msmtdemultiscalesemanticsminingandtinydetailsenhancementnetworkforretinalvesselsegmentation AT xiongli msmtdemultiscalesemanticsminingandtinydetailsenhancementnetworkforretinalvesselsegmentation AT jingqinlv msmtdemultiscalesemanticsminingandtinydetailsenhancementnetworkforretinalvesselsegmentation AT donghongji msmtdemultiscalesemanticsminingandtinydetailsenhancementnetworkforretinalvesselsegmentation |