Loading…

Spatial attention U-Net model with Harris hawks optimization for retinal blood vessel and optic disc segmentation in fundus images

Background The state of the human eye’s blood vessels is a crucial aspect in the diagnosis of ophthalmological illnesses. For many computer-aided diagnostic systems, precise retinal vessel segmentation is an essential job. However, it remains a difficult task due to the intricate vascular system of...

Full description

Saved in:
Bibliographic Details
Published in:International ophthalmology 2024-08, Vol.44 (1), p.359, Article 359
Main Authors: Kumar, Puranam Revanth, Shilpa, B., Jha, Rajesh Kumar, Chellibouina, Veni Sree
Format: Article
Language:English
Subjects:
Citations: Items that this one cites
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
cited_by
cites cdi_FETCH-LOGICAL-c256t-d9904bbdbd166572ff0e6f0548ef35336b5b4ee6058d5ee53c48f4bb5d9537693
container_end_page
container_issue 1
container_start_page 359
container_title International ophthalmology
container_volume 44
creator Kumar, Puranam Revanth
Shilpa, B.
Jha, Rajesh Kumar
Chellibouina, Veni Sree
description Background The state of the human eye’s blood vessels is a crucial aspect in the diagnosis of ophthalmological illnesses. For many computer-aided diagnostic systems, precise retinal vessel segmentation is an essential job. However, it remains a difficult task due to the intricate vascular system of the eye. Although many different vascular segmentation techniques have already been presented, additional study is still required to address the problem of inadequate segmentation of thin and tiny vessels. Methods In this work, we introduce the Spatial Attention U-Net (SAU-Net) model with harris hawks’ optimization (HHO), a lightweight network that can be applied as a data augmentation technique to improve the efficiency of the existing annotated samples without the need of thousands of training instances for Retinal Blood Vessel and Optic Disc Segmentation. The SAU-Net-HHO implementation uses a spatially inferred attention map multiplied by the input feature map for adaptive feature enhancement. U-Net convolutional blocks have been replaced with structured dropout blocks in the proposed network to prevent overfitting. Data from both vascular extraction (DRIVE) and structured analysis of the retina (STARE) are used to evaluate SAU-Net-HHO performance. Results The results show that the proposed SAU-Net-HHO performs well on both datasets. Analysing the obtained results, an average of 98.5% accuracy and Specificity 96.7% was achieved for DRIVE dataset and 97.8% accuracy and specificity 94.5% for STARE dataset. The proposed method yields numerical results with average values that are on par with those of state-of-the-art methods. Conclusion Visual inspection has revealed that the suggested method can segment thin and tiny vessels with greater accuracy than previous methods. It also demonstrates its potential for real-life clinical application.
doi_str_mv 10.1007/s10792-024-03279-3
format article
fullrecord <record><control><sourceid>proquest_cross</sourceid><recordid>TN_cdi_proquest_miscellaneous_3099799688</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>3099799688</sourcerecordid><originalsourceid>FETCH-LOGICAL-c256t-d9904bbdbd166572ff0e6f0548ef35336b5b4ee6058d5ee53c48f4bb5d9537693</originalsourceid><addsrcrecordid>eNp9kT1v1TAUhq0K1O8_0AFZYmEJnMSxE49VBRSpggE6W058cuuSxLc-DhWM_PKamxYQA5Mt-Xle-_hl7KyE1yVA84ZKaHRVQFUXIKpGF2KPHZayEUWlBDz7a3_AjohuAUA3Wu2zA6EraFQtD9nPz1ubvB25TQnn5MPMr4uPmPgUHI783qcbfmlj9MRv7P1X4mGb_OR_2B06hMgjJj_ngG4MwfFvSJQ9O7sd2XPnqeeEmymnr5LP3jK7hbif7AbphD0f7Eh4-rges-t3b79cXBZXn95_uDi_KvpKqlQ4raHuOte5UinZVMMAqAaQdYuDkEKoTnY1ogLZOokoRV-3Qxak01I0Sotj9mrN3cZwtyAlM-W34TjaGcNCRoDO36NV22b05T_obVhinnJHtXUtpC4zVa1UHwNRxMFsYx4pfjclmF8NmbUhkxsyu4aMyNKLx-ilm9D9Vp4qyYBYAcpH8wbjn7v_E_sA6PKddQ</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>3098443591</pqid></control><display><type>article</type><title>Spatial attention U-Net model with Harris hawks optimization for retinal blood vessel and optic disc segmentation in fundus images</title><source>Springer Link</source><creator>Kumar, Puranam Revanth ; Shilpa, B. ; Jha, Rajesh Kumar ; Chellibouina, Veni Sree</creator><creatorcontrib>Kumar, Puranam Revanth ; Shilpa, B. ; Jha, Rajesh Kumar ; Chellibouina, Veni Sree</creatorcontrib><description>Background The state of the human eye’s blood vessels is a crucial aspect in the diagnosis of ophthalmological illnesses. For many computer-aided diagnostic systems, precise retinal vessel segmentation is an essential job. However, it remains a difficult task due to the intricate vascular system of the eye. Although many different vascular segmentation techniques have already been presented, additional study is still required to address the problem of inadequate segmentation of thin and tiny vessels. Methods In this work, we introduce the Spatial Attention U-Net (SAU-Net) model with harris hawks’ optimization (HHO), a lightweight network that can be applied as a data augmentation technique to improve the efficiency of the existing annotated samples without the need of thousands of training instances for Retinal Blood Vessel and Optic Disc Segmentation. The SAU-Net-HHO implementation uses a spatially inferred attention map multiplied by the input feature map for adaptive feature enhancement. U-Net convolutional blocks have been replaced with structured dropout blocks in the proposed network to prevent overfitting. Data from both vascular extraction (DRIVE) and structured analysis of the retina (STARE) are used to evaluate SAU-Net-HHO performance. Results The results show that the proposed SAU-Net-HHO performs well on both datasets. Analysing the obtained results, an average of 98.5% accuracy and Specificity 96.7% was achieved for DRIVE dataset and 97.8% accuracy and specificity 94.5% for STARE dataset. The proposed method yields numerical results with average values that are on par with those of state-of-the-art methods. Conclusion Visual inspection has revealed that the suggested method can segment thin and tiny vessels with greater accuracy than previous methods. It also demonstrates its potential for real-life clinical application.</description><identifier>ISSN: 1573-2630</identifier><identifier>ISSN: 0165-5701</identifier><identifier>EISSN: 1573-2630</identifier><identifier>DOI: 10.1007/s10792-024-03279-3</identifier><identifier>PMID: 39207645</identifier><language>eng</language><publisher>Dordrecht: Springer Netherlands</publisher><subject>Accuracy ; Adaptive sampling ; Algorithms ; Blood vessels ; Data augmentation ; Datasets ; Diagnostic systems ; Eye ; Eye (anatomy) ; Feature maps ; Fundus Oculi ; Humans ; Image processing ; Image segmentation ; Medicine ; Medicine &amp; Public Health ; Neural Networks, Computer ; Ophthalmology ; Optic Disk - blood supply ; Optic Disk - diagnostic imaging ; Optimization ; Original Paper ; Retina ; Retinal Vessels - diagnostic imaging ; Segmentation ; Vascular system</subject><ispartof>International ophthalmology, 2024-08, Vol.44 (1), p.359, Article 359</ispartof><rights>The Author(s), under exclusive licence to Springer Nature B.V. 2024. Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.</rights><rights>2024. The Author(s), under exclusive licence to Springer Nature B.V.</rights><lds50>peer_reviewed</lds50><woscitedreferencessubscribed>false</woscitedreferencessubscribed><cites>FETCH-LOGICAL-c256t-d9904bbdbd166572ff0e6f0548ef35336b5b4ee6058d5ee53c48f4bb5d9537693</cites></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>314,780,784,27924,27925</link.rule.ids><backlink>$$Uhttps://www.ncbi.nlm.nih.gov/pubmed/39207645$$D View this record in MEDLINE/PubMed$$Hfree_for_read</backlink></links><search><creatorcontrib>Kumar, Puranam Revanth</creatorcontrib><creatorcontrib>Shilpa, B.</creatorcontrib><creatorcontrib>Jha, Rajesh Kumar</creatorcontrib><creatorcontrib>Chellibouina, Veni Sree</creatorcontrib><title>Spatial attention U-Net model with Harris hawks optimization for retinal blood vessel and optic disc segmentation in fundus images</title><title>International ophthalmology</title><addtitle>Int Ophthalmol</addtitle><addtitle>Int Ophthalmol</addtitle><description>Background The state of the human eye’s blood vessels is a crucial aspect in the diagnosis of ophthalmological illnesses. For many computer-aided diagnostic systems, precise retinal vessel segmentation is an essential job. However, it remains a difficult task due to the intricate vascular system of the eye. Although many different vascular segmentation techniques have already been presented, additional study is still required to address the problem of inadequate segmentation of thin and tiny vessels. Methods In this work, we introduce the Spatial Attention U-Net (SAU-Net) model with harris hawks’ optimization (HHO), a lightweight network that can be applied as a data augmentation technique to improve the efficiency of the existing annotated samples without the need of thousands of training instances for Retinal Blood Vessel and Optic Disc Segmentation. The SAU-Net-HHO implementation uses a spatially inferred attention map multiplied by the input feature map for adaptive feature enhancement. U-Net convolutional blocks have been replaced with structured dropout blocks in the proposed network to prevent overfitting. Data from both vascular extraction (DRIVE) and structured analysis of the retina (STARE) are used to evaluate SAU-Net-HHO performance. Results The results show that the proposed SAU-Net-HHO performs well on both datasets. Analysing the obtained results, an average of 98.5% accuracy and Specificity 96.7% was achieved for DRIVE dataset and 97.8% accuracy and specificity 94.5% for STARE dataset. The proposed method yields numerical results with average values that are on par with those of state-of-the-art methods. Conclusion Visual inspection has revealed that the suggested method can segment thin and tiny vessels with greater accuracy than previous methods. It also demonstrates its potential for real-life clinical application.</description><subject>Accuracy</subject><subject>Adaptive sampling</subject><subject>Algorithms</subject><subject>Blood vessels</subject><subject>Data augmentation</subject><subject>Datasets</subject><subject>Diagnostic systems</subject><subject>Eye</subject><subject>Eye (anatomy)</subject><subject>Feature maps</subject><subject>Fundus Oculi</subject><subject>Humans</subject><subject>Image processing</subject><subject>Image segmentation</subject><subject>Medicine</subject><subject>Medicine &amp; Public Health</subject><subject>Neural Networks, Computer</subject><subject>Ophthalmology</subject><subject>Optic Disk - blood supply</subject><subject>Optic Disk - diagnostic imaging</subject><subject>Optimization</subject><subject>Original Paper</subject><subject>Retina</subject><subject>Retinal Vessels - diagnostic imaging</subject><subject>Segmentation</subject><subject>Vascular system</subject><issn>1573-2630</issn><issn>0165-5701</issn><issn>1573-2630</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2024</creationdate><recordtype>article</recordtype><recordid>eNp9kT1v1TAUhq0K1O8_0AFZYmEJnMSxE49VBRSpggE6W058cuuSxLc-DhWM_PKamxYQA5Mt-Xle-_hl7KyE1yVA84ZKaHRVQFUXIKpGF2KPHZayEUWlBDz7a3_AjohuAUA3Wu2zA6EraFQtD9nPz1ubvB25TQnn5MPMr4uPmPgUHI783qcbfmlj9MRv7P1X4mGb_OR_2B06hMgjJj_ngG4MwfFvSJQ9O7sd2XPnqeeEmymnr5LP3jK7hbif7AbphD0f7Eh4-rges-t3b79cXBZXn95_uDi_KvpKqlQ4raHuOte5UinZVMMAqAaQdYuDkEKoTnY1ogLZOokoRV-3Qxak01I0Sotj9mrN3cZwtyAlM-W34TjaGcNCRoDO36NV22b05T_obVhinnJHtXUtpC4zVa1UHwNRxMFsYx4pfjclmF8NmbUhkxsyu4aMyNKLx-ilm9D9Vp4qyYBYAcpH8wbjn7v_E_sA6PKddQ</recordid><startdate>20240829</startdate><enddate>20240829</enddate><creator>Kumar, Puranam Revanth</creator><creator>Shilpa, B.</creator><creator>Jha, Rajesh Kumar</creator><creator>Chellibouina, Veni Sree</creator><general>Springer Netherlands</general><general>Springer Nature B.V</general><scope>CGR</scope><scope>CUY</scope><scope>CVF</scope><scope>ECM</scope><scope>EIF</scope><scope>NPM</scope><scope>AAYXX</scope><scope>CITATION</scope><scope>7QL</scope><scope>7T7</scope><scope>7TK</scope><scope>7U9</scope><scope>8FD</scope><scope>C1K</scope><scope>FR3</scope><scope>H94</scope><scope>K9.</scope><scope>M7N</scope><scope>P64</scope><scope>7X8</scope></search><sort><creationdate>20240829</creationdate><title>Spatial attention U-Net model with Harris hawks optimization for retinal blood vessel and optic disc segmentation in fundus images</title><author>Kumar, Puranam Revanth ; Shilpa, B. ; Jha, Rajesh Kumar ; Chellibouina, Veni Sree</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c256t-d9904bbdbd166572ff0e6f0548ef35336b5b4ee6058d5ee53c48f4bb5d9537693</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2024</creationdate><topic>Accuracy</topic><topic>Adaptive sampling</topic><topic>Algorithms</topic><topic>Blood vessels</topic><topic>Data augmentation</topic><topic>Datasets</topic><topic>Diagnostic systems</topic><topic>Eye</topic><topic>Eye (anatomy)</topic><topic>Feature maps</topic><topic>Fundus Oculi</topic><topic>Humans</topic><topic>Image processing</topic><topic>Image segmentation</topic><topic>Medicine</topic><topic>Medicine &amp; Public Health</topic><topic>Neural Networks, Computer</topic><topic>Ophthalmology</topic><topic>Optic Disk - blood supply</topic><topic>Optic Disk - diagnostic imaging</topic><topic>Optimization</topic><topic>Original Paper</topic><topic>Retina</topic><topic>Retinal Vessels - diagnostic imaging</topic><topic>Segmentation</topic><topic>Vascular system</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Kumar, Puranam Revanth</creatorcontrib><creatorcontrib>Shilpa, B.</creatorcontrib><creatorcontrib>Jha, Rajesh Kumar</creatorcontrib><creatorcontrib>Chellibouina, Veni Sree</creatorcontrib><collection>Medline</collection><collection>MEDLINE</collection><collection>MEDLINE (Ovid)</collection><collection>MEDLINE</collection><collection>MEDLINE</collection><collection>PubMed</collection><collection>CrossRef</collection><collection>Bacteriology Abstracts (Microbiology B)</collection><collection>Industrial and Applied Microbiology Abstracts (Microbiology A)</collection><collection>Neurosciences Abstracts</collection><collection>Virology and AIDS Abstracts</collection><collection>Technology Research Database</collection><collection>Environmental Sciences and Pollution Management</collection><collection>Engineering Research Database</collection><collection>AIDS and Cancer Research Abstracts</collection><collection>ProQuest Health &amp; Medical Complete (Alumni)</collection><collection>Algology Mycology and Protozoology Abstracts (Microbiology C)</collection><collection>Biotechnology and BioEngineering Abstracts</collection><collection>MEDLINE - Academic</collection><jtitle>International ophthalmology</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Kumar, Puranam Revanth</au><au>Shilpa, B.</au><au>Jha, Rajesh Kumar</au><au>Chellibouina, Veni Sree</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Spatial attention U-Net model with Harris hawks optimization for retinal blood vessel and optic disc segmentation in fundus images</atitle><jtitle>International ophthalmology</jtitle><stitle>Int Ophthalmol</stitle><addtitle>Int Ophthalmol</addtitle><date>2024-08-29</date><risdate>2024</risdate><volume>44</volume><issue>1</issue><spage>359</spage><pages>359-</pages><artnum>359</artnum><issn>1573-2630</issn><issn>0165-5701</issn><eissn>1573-2630</eissn><abstract>Background The state of the human eye’s blood vessels is a crucial aspect in the diagnosis of ophthalmological illnesses. For many computer-aided diagnostic systems, precise retinal vessel segmentation is an essential job. However, it remains a difficult task due to the intricate vascular system of the eye. Although many different vascular segmentation techniques have already been presented, additional study is still required to address the problem of inadequate segmentation of thin and tiny vessels. Methods In this work, we introduce the Spatial Attention U-Net (SAU-Net) model with harris hawks’ optimization (HHO), a lightweight network that can be applied as a data augmentation technique to improve the efficiency of the existing annotated samples without the need of thousands of training instances for Retinal Blood Vessel and Optic Disc Segmentation. The SAU-Net-HHO implementation uses a spatially inferred attention map multiplied by the input feature map for adaptive feature enhancement. U-Net convolutional blocks have been replaced with structured dropout blocks in the proposed network to prevent overfitting. Data from both vascular extraction (DRIVE) and structured analysis of the retina (STARE) are used to evaluate SAU-Net-HHO performance. Results The results show that the proposed SAU-Net-HHO performs well on both datasets. Analysing the obtained results, an average of 98.5% accuracy and Specificity 96.7% was achieved for DRIVE dataset and 97.8% accuracy and specificity 94.5% for STARE dataset. The proposed method yields numerical results with average values that are on par with those of state-of-the-art methods. Conclusion Visual inspection has revealed that the suggested method can segment thin and tiny vessels with greater accuracy than previous methods. It also demonstrates its potential for real-life clinical application.</abstract><cop>Dordrecht</cop><pub>Springer Netherlands</pub><pmid>39207645</pmid><doi>10.1007/s10792-024-03279-3</doi></addata></record>
fulltext fulltext
identifier ISSN: 1573-2630
ispartof International ophthalmology, 2024-08, Vol.44 (1), p.359, Article 359
issn 1573-2630
0165-5701
1573-2630
language eng
recordid cdi_proquest_miscellaneous_3099799688
source Springer Link
subjects Accuracy
Adaptive sampling
Algorithms
Blood vessels
Data augmentation
Datasets
Diagnostic systems
Eye
Eye (anatomy)
Feature maps
Fundus Oculi
Humans
Image processing
Image segmentation
Medicine
Medicine & Public Health
Neural Networks, Computer
Ophthalmology
Optic Disk - blood supply
Optic Disk - diagnostic imaging
Optimization
Original Paper
Retina
Retinal Vessels - diagnostic imaging
Segmentation
Vascular system
title Spatial attention U-Net model with Harris hawks optimization for retinal blood vessel and optic disc segmentation in fundus images
url http://sfxeu10.hosted.exlibrisgroup.com/loughborough?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-23T13%3A46%3A51IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_cross&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Spatial%20attention%20U-Net%20model%20with%20Harris%20hawks%20optimization%20for%20retinal%20blood%20vessel%20and%20optic%20disc%20segmentation%20in%20fundus%20images&rft.jtitle=International%20ophthalmology&rft.au=Kumar,%20Puranam%20Revanth&rft.date=2024-08-29&rft.volume=44&rft.issue=1&rft.spage=359&rft.pages=359-&rft.artnum=359&rft.issn=1573-2630&rft.eissn=1573-2630&rft_id=info:doi/10.1007/s10792-024-03279-3&rft_dat=%3Cproquest_cross%3E3099799688%3C/proquest_cross%3E%3Cgrp_id%3Ecdi_FETCH-LOGICAL-c256t-d9904bbdbd166572ff0e6f0548ef35336b5b4ee6058d5ee53c48f4bb5d9537693%3C/grp_id%3E%3Coa%3E%3C/oa%3E%3Curl%3E%3C/url%3E&rft_id=info:oai/&rft_pqid=3098443591&rft_id=info:pmid/39207645&rfr_iscdi=true