EAMR-Net: A multiscale effective spatial and cross-channel attention network for retinal vessel segmentation
Delineation of retinal vessels in fundus images is essential for detecting a range of eye disorders. An automated technique for vessel segmentation can assist clinicians and enhance the efficiency of the diagnostic process. Traditional methods fail to extract multiscale information, discard unnecess...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
AIMS Press
2024-02-01
|
| Series: | Mathematical Biosciences and Engineering |
| Subjects: | |
| Online Access: | https://www.aimspress.com/article/doi/10.3934/mbe.2024208?viewType=HTML |
| _version_ | 1797256722379702272 |
|---|---|
| author | G. Prethija Jeevaa Katiravan |
| author_facet | G. Prethija Jeevaa Katiravan |
| author_sort | G. Prethija |
| collection | DOAJ |
| description | Delineation of retinal vessels in fundus images is essential for detecting a range of eye disorders. An automated technique for vessel segmentation can assist clinicians and enhance the efficiency of the diagnostic process. Traditional methods fail to extract multiscale information, discard unnecessary information, and delineate thin vessels. In this paper, a novel residual U-Net architecture that incorporates multi-scale feature learning and effective attention is proposed to delineate the retinal vessels precisely. Since drop block regularization performs better than drop out in preventing overfitting, drop block was used in this study. A multi-scale feature learning module was added instead of a skip connection to learn multi-scale features. A novel effective attention block was proposed and integrated with the decoder block to obtain precise spatial and channel information. Experimental findings indicated that the proposed model exhibited outstanding performance in retinal vessel delineation. The sensitivities achieved for DRIVE, STARE, and CHASE_DB datasets were 0.8293, 0.8151 and 0.8084, respectively. |
| first_indexed | 2024-04-24T22:26:16Z |
| format | Article |
| id | doaj.art-98473079fa4b4b8bbb5453b409341f21 |
| institution | Directory Open Access Journal |
| issn | 1551-0018 |
| language | English |
| last_indexed | 2024-04-24T22:26:16Z |
| publishDate | 2024-02-01 |
| publisher | AIMS Press |
| record_format | Article |
| series | Mathematical Biosciences and Engineering |
| spelling | doaj.art-98473079fa4b4b8bbb5453b409341f212024-03-20T01:25:35ZengAIMS PressMathematical Biosciences and Engineering1551-00182024-02-012134742476110.3934/mbe.2024208EAMR-Net: A multiscale effective spatial and cross-channel attention network for retinal vessel segmentationG. Prethija0Jeevaa Katiravan11. School of Computer Science and Engineering, Vellore Institute of Technology, Chennai 600127, India2. Department of Information Technology, Velammal Engineering College, Chennai 600066, IndiaDelineation of retinal vessels in fundus images is essential for detecting a range of eye disorders. An automated technique for vessel segmentation can assist clinicians and enhance the efficiency of the diagnostic process. Traditional methods fail to extract multiscale information, discard unnecessary information, and delineate thin vessels. In this paper, a novel residual U-Net architecture that incorporates multi-scale feature learning and effective attention is proposed to delineate the retinal vessels precisely. Since drop block regularization performs better than drop out in preventing overfitting, drop block was used in this study. A multi-scale feature learning module was added instead of a skip connection to learn multi-scale features. A novel effective attention block was proposed and integrated with the decoder block to obtain precise spatial and channel information. Experimental findings indicated that the proposed model exhibited outstanding performance in retinal vessel delineation. The sensitivities achieved for DRIVE, STARE, and CHASE_DB datasets were 0.8293, 0.8151 and 0.8084, respectively.https://www.aimspress.com/article/doi/10.3934/mbe.2024208?viewType=HTMLresidualattentiondrop blockspatial poolingu-net |
| spellingShingle | G. Prethija Jeevaa Katiravan EAMR-Net: A multiscale effective spatial and cross-channel attention network for retinal vessel segmentation Mathematical Biosciences and Engineering residual attention drop block spatial pooling u-net |
| title | EAMR-Net: A multiscale effective spatial and cross-channel attention network for retinal vessel segmentation |
| title_full | EAMR-Net: A multiscale effective spatial and cross-channel attention network for retinal vessel segmentation |
| title_fullStr | EAMR-Net: A multiscale effective spatial and cross-channel attention network for retinal vessel segmentation |
| title_full_unstemmed | EAMR-Net: A multiscale effective spatial and cross-channel attention network for retinal vessel segmentation |
| title_short | EAMR-Net: A multiscale effective spatial and cross-channel attention network for retinal vessel segmentation |
| title_sort | eamr net a multiscale effective spatial and cross channel attention network for retinal vessel segmentation |
| topic | residual attention drop block spatial pooling u-net |
| url | https://www.aimspress.com/article/doi/10.3934/mbe.2024208?viewType=HTML |
| work_keys_str_mv | AT gprethija eamrnetamultiscaleeffectivespatialandcrosschannelattentionnetworkforretinalvesselsegmentation AT jeevaakatiravan eamrnetamultiscaleeffectivespatialandcrosschannelattentionnetworkforretinalvesselsegmentation |