An effective vessel segmentation method using SLOA-HGC
Abstract Accurate segmentation of retinal blood vessels from retinal images is crucial for detecting and diagnosing a wide range of ophthalmic diseases. Our retinal blood vessel segmentation algorithm enhances microfine vessel extraction, improves edge texture clarity, and normalizes vessel distribu...
Saved in:
| Main Authors: | , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2025-01-01
|
| Series: | Scientific Reports |
| Subjects: | |
| Online Access: | https://doi.org/10.1038/s41598-024-84901-3 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841544659626622976 |
|---|---|
| author | Zerui Liu Junliang Du Weisi Dai Wenke Zhu Ziqing Ye Lin Li Zewei Liu Linan Hu Lin Chen Lixiang Sun |
| author_facet | Zerui Liu Junliang Du Weisi Dai Wenke Zhu Ziqing Ye Lin Li Zewei Liu Linan Hu Lin Chen Lixiang Sun |
| author_sort | Zerui Liu |
| collection | DOAJ |
| description | Abstract Accurate segmentation of retinal blood vessels from retinal images is crucial for detecting and diagnosing a wide range of ophthalmic diseases. Our retinal blood vessel segmentation algorithm enhances microfine vessel extraction, improves edge texture clarity, and normalizes vessel distribution. It stabilizes neural network training for complex retinal vascular features. Channel-aware self-attention (CAS) improves microfine vessel segmentation sensitivity. Heterogeneous adaptive pooling (HAP) facilitates accurate vessel edge segmentation through multi-scale feature extraction. The ghost fully convolutional Rectified Linear Unit (GFCReLU) module in the output convolutional layer captures deep semantic information for better vessel localization. Optimization training with Sparrow-Integrated Lion Optimization Algorithm (SLOA) employs sparrow stochastic updating and annealing to fine-tune parameters. The results of the experiments on our homemade dataset and three public datasets are as follows: Mean Intersection over Union (MIoU) of 80.61%, 76.14%, 76.90%, 74.11%; Dice coefficients of 78.97%, 72.51%, 72.84%, 68.93%; and accuracies of 94.83%, 95.74%, 96.67%, 95.81% respectively. The model effectively segments retinal blood vessels, offering potential for diagnosing ophthalmic diseases. Our dataset is available at https://github.com/ZhouGuoXiong/Retinal-blood-vessels-for-segmentation . |
| format | Article |
| id | doaj-art-4da3188b06f64dc8a296fdec8c683e6f |
| institution | Kabale University |
| issn | 2045-2322 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Scientific Reports |
| spelling | doaj-art-4da3188b06f64dc8a296fdec8c683e6f2025-01-12T12:21:59ZengNature PortfolioScientific Reports2045-23222025-01-0115112210.1038/s41598-024-84901-3An effective vessel segmentation method using SLOA-HGCZerui Liu0Junliang Du1Weisi Dai2Wenke Zhu3Ziqing Ye4Lin Li5Zewei Liu6Linan Hu7Lin Chen8Lixiang Sun9Faculty of Electronic Information and Physics, Central South University of Forestry and TechnologyArtificial Intelligence Institute, Shanghai Jiao Tong UniversityFaculty of Electronic Information and Physics, Central South University of Forestry and TechnologyCollege of Bangor, Central South University of Forestry and TechnologyFaculty of Electronic Information and Physics, Central South University of Forestry and TechnologyFaculty of Electronic Information and Physics, Central South University of Forestry and TechnologyFaculty of Electronic Information and Physics, Central South University of Forestry and TechnologyZhuzhou Central HospitalZhuzhou Sansanyi Eye HospitalFaculty of Electronic Information and Physics, Central South University of Forestry and TechnologyAbstract Accurate segmentation of retinal blood vessels from retinal images is crucial for detecting and diagnosing a wide range of ophthalmic diseases. Our retinal blood vessel segmentation algorithm enhances microfine vessel extraction, improves edge texture clarity, and normalizes vessel distribution. It stabilizes neural network training for complex retinal vascular features. Channel-aware self-attention (CAS) improves microfine vessel segmentation sensitivity. Heterogeneous adaptive pooling (HAP) facilitates accurate vessel edge segmentation through multi-scale feature extraction. The ghost fully convolutional Rectified Linear Unit (GFCReLU) module in the output convolutional layer captures deep semantic information for better vessel localization. Optimization training with Sparrow-Integrated Lion Optimization Algorithm (SLOA) employs sparrow stochastic updating and annealing to fine-tune parameters. The results of the experiments on our homemade dataset and three public datasets are as follows: Mean Intersection over Union (MIoU) of 80.61%, 76.14%, 76.90%, 74.11%; Dice coefficients of 78.97%, 72.51%, 72.84%, 68.93%; and accuracies of 94.83%, 95.74%, 96.67%, 95.81% respectively. The model effectively segments retinal blood vessels, offering potential for diagnosing ophthalmic diseases. Our dataset is available at https://github.com/ZhouGuoXiong/Retinal-blood-vessels-for-segmentation .https://doi.org/10.1038/s41598-024-84901-3Vessel segmentationSparrow updating mechanismArtificial intelligence applied medical image processing |
| spellingShingle | Zerui Liu Junliang Du Weisi Dai Wenke Zhu Ziqing Ye Lin Li Zewei Liu Linan Hu Lin Chen Lixiang Sun An effective vessel segmentation method using SLOA-HGC Scientific Reports Vessel segmentation Sparrow updating mechanism Artificial intelligence applied medical image processing |
| title | An effective vessel segmentation method using SLOA-HGC |
| title_full | An effective vessel segmentation method using SLOA-HGC |
| title_fullStr | An effective vessel segmentation method using SLOA-HGC |
| title_full_unstemmed | An effective vessel segmentation method using SLOA-HGC |
| title_short | An effective vessel segmentation method using SLOA-HGC |
| title_sort | effective vessel segmentation method using sloa hgc |
| topic | Vessel segmentation Sparrow updating mechanism Artificial intelligence applied medical image processing |
| url | https://doi.org/10.1038/s41598-024-84901-3 |
| work_keys_str_mv | AT zeruiliu aneffectivevesselsegmentationmethodusingsloahgc AT junliangdu aneffectivevesselsegmentationmethodusingsloahgc AT weisidai aneffectivevesselsegmentationmethodusingsloahgc AT wenkezhu aneffectivevesselsegmentationmethodusingsloahgc AT ziqingye aneffectivevesselsegmentationmethodusingsloahgc AT linli aneffectivevesselsegmentationmethodusingsloahgc AT zeweiliu aneffectivevesselsegmentationmethodusingsloahgc AT linanhu aneffectivevesselsegmentationmethodusingsloahgc AT linchen aneffectivevesselsegmentationmethodusingsloahgc AT lixiangsun aneffectivevesselsegmentationmethodusingsloahgc AT zeruiliu effectivevesselsegmentationmethodusingsloahgc AT junliangdu effectivevesselsegmentationmethodusingsloahgc AT weisidai effectivevesselsegmentationmethodusingsloahgc AT wenkezhu effectivevesselsegmentationmethodusingsloahgc AT ziqingye effectivevesselsegmentationmethodusingsloahgc AT linli effectivevesselsegmentationmethodusingsloahgc AT zeweiliu effectivevesselsegmentationmethodusingsloahgc AT linanhu effectivevesselsegmentationmethodusingsloahgc AT linchen effectivevesselsegmentationmethodusingsloahgc AT lixiangsun effectivevesselsegmentationmethodusingsloahgc |