Harmonizing Unets: Attention Fusion module in cascaded-Unets for low-quality OCT image fluid segmentation
Optical coherence tomography (OCT) is widely used for its high resolution. Accurate OCT image segmentation can significantly improve the diagnosis and treatment of retinal diseases such as Diabetic Macular Edema (DME). However, in resource-limited regions, portable devices with low-quality output ar...
Gespeichert in:
| Veröffentlicht in: | Computers in biology and medicine 2024-12, Vol.183, p.109223, Article 109223 |
|---|---|
| Hauptverfasser: | , , , , , , , , |
| Format: | Artikel |
| Sprache: | eng |
| Schlagworte: | |
| Online-Zugang: | Volltext |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| container_end_page | |
|---|---|
| container_issue | |
| container_start_page | 109223 |
| container_title | Computers in biology and medicine |
| container_volume | 183 |
| creator | Wu, Zhuoyu Wu, Qinchen Fang, Wenqi Ou, Wenhui Wang, Quanjun Zhang, Linde Chen, Chao Wang, Zheng Li, Heshan |
| description | Optical coherence tomography (OCT) is widely used for its high resolution. Accurate OCT image segmentation can significantly improve the diagnosis and treatment of retinal diseases such as Diabetic Macular Edema (DME). However, in resource-limited regions, portable devices with low-quality output are more frequently used, severely affecting the performance of segmentation. To address this issue, we propose a novel methodology in this paper, including a dedicated pre-processing pipeline and an end-to-end double U-shaped cascaded architecture, H-Unets. In addition, an Adaptive Attention Fusion (AAF) module is elaborately designed to improve the segmentation performance of H-Unets. To demonstrate the effectiveness of our method, we conduct a bunch of ablation and comparative studies on three open-source datasets. The experimental results show the validity of the pre-processing pipeline and H-Unets, achieving the highest Dice score of 90.60%±0.87% among popular methods in a relatively small model size. |
| doi_str_mv | 10.1016/j.compbiomed.2024.109223 |
| format | Article |
| fullrecord | <record><control><sourceid>proquest_cross</sourceid><recordid>TN_cdi_proquest_miscellaneous_3113381430</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><els_id>S0010482524013088</els_id><sourcerecordid>3113381430</sourcerecordid><originalsourceid>FETCH-LOGICAL-c1922-de8480eda524b267bcc7f1414e7f9a3521ecbb89047f60997e9841721ac616283</originalsourceid><addsrcrecordid>eNqFkUFv1DAQhS0EotvCX0CWuHDJ4rGdxOFWVrRFqtRLe7Yce7LyKrG3dgIqvx6HbYXEhdNI9jdvZt4jhALbAoPm82Fr43TsfZzQbTnjsjx3nItXZAOq7SpWC_mabBgDVknF6zNynvOBMSaZYG_JmehEowTwDfE3Jk0x-F8-7OlDwDl_oZfzjGH2MdCrJa9lim4ZkfpArcnWOHTVH5QOMdEx_qweFzP6-Yne7e6pn8we6TAu3tGM-6lImVXsHXkzmDHj--d6QR6uvt3vbqrbu-vvu8vbykI5oXKopGLoTM1lz5u2t7YdQILEduiMqDmg7XvVMdkODeu6FjsloeVgbAMNV-KCfDrpHlN8XDDPevLZ4jiagHHJWgAIoUAKVtCP_6CHuKRQtisUL77VAtpCqRNlU8w54aCPqRyZnjQwvcahD_pvHHqNQ5_iKK0fngcs_fr30vjifwG-ngAsjvzwmHS2HoNF5xPaWbvo_z_lN6s9n9I</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>3128255317</pqid></control><display><type>article</type><title>Harmonizing Unets: Attention Fusion module in cascaded-Unets for low-quality OCT image fluid segmentation</title><source>MEDLINE</source><source>ScienceDirect Journals (5 years ago - present)</source><creator>Wu, Zhuoyu ; Wu, Qinchen ; Fang, Wenqi ; Ou, Wenhui ; Wang, Quanjun ; Zhang, Linde ; Chen, Chao ; Wang, Zheng ; Li, Heshan</creator><creatorcontrib>Wu, Zhuoyu ; Wu, Qinchen ; Fang, Wenqi ; Ou, Wenhui ; Wang, Quanjun ; Zhang, Linde ; Chen, Chao ; Wang, Zheng ; Li, Heshan</creatorcontrib><description>Optical coherence tomography (OCT) is widely used for its high resolution. Accurate OCT image segmentation can significantly improve the diagnosis and treatment of retinal diseases such as Diabetic Macular Edema (DME). However, in resource-limited regions, portable devices with low-quality output are more frequently used, severely affecting the performance of segmentation. To address this issue, we propose a novel methodology in this paper, including a dedicated pre-processing pipeline and an end-to-end double U-shaped cascaded architecture, H-Unets. In addition, an Adaptive Attention Fusion (AAF) module is elaborately designed to improve the segmentation performance of H-Unets. To demonstrate the effectiveness of our method, we conduct a bunch of ablation and comparative studies on three open-source datasets. The experimental results show the validity of the pre-processing pipeline and H-Unets, achieving the highest Dice score of 90.60%±0.87% among popular methods in a relatively small model size.</description><identifier>ISSN: 0010-4825</identifier><identifier>ISSN: 1879-0534</identifier><identifier>EISSN: 1879-0534</identifier><identifier>DOI: 10.1016/j.compbiomed.2024.109223</identifier><identifier>PMID: 39368312</identifier><language>eng</language><publisher>United States: Elsevier Ltd</publisher><subject>Ablation ; Accuracy ; Algorithms ; Artificial intelligence ; Biology ; Comparative studies ; Computers ; Datasets ; Deep learning ; Diabetes ; Diabetes mellitus ; Diabetic Retinopathy - diagnostic imaging ; Edema ; Efficiency ; Fusion module ; Humans ; Image processing ; Image Processing, Computer-Assisted - methods ; Image quality ; Image resolution ; Image segmentation ; Intraretinal fluid detection ; Low income groups ; Macular Edema - diagnostic imaging ; Medical image segmentation ; Medical imaging ; Medicine ; Modules ; Optical Coherence Tomography ; Portable equipment ; Public health ; Tomography ; Tomography, Optical Coherence - methods</subject><ispartof>Computers in biology and medicine, 2024-12, Vol.183, p.109223, Article 109223</ispartof><rights>2024 Elsevier Ltd</rights><rights>Copyright © 2024 Elsevier Ltd. All rights reserved.</rights><rights>2024. Elsevier Ltd</rights><lds50>peer_reviewed</lds50><woscitedreferencessubscribed>false</woscitedreferencessubscribed><cites>FETCH-LOGICAL-c1922-de8480eda524b267bcc7f1414e7f9a3521ecbb89047f60997e9841721ac616283</cites><orcidid>0009-0009-6779-5841 ; 0009-0002-1978-8021 ; 0009-0002-5147-4858 ; 0000-0001-6488-224X ; 0000-0003-3347-0511 ; 0000-0002-9075-7626</orcidid></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktohtml>$$Uhttps://dx.doi.org/10.1016/j.compbiomed.2024.109223$$EHTML$$P50$$Gelsevier$$H</linktohtml><link.rule.ids>314,780,784,3550,27924,27925,45995</link.rule.ids><backlink>$$Uhttps://www.ncbi.nlm.nih.gov/pubmed/39368312$$D View this record in MEDLINE/PubMed$$Hfree_for_read</backlink></links><search><creatorcontrib>Wu, Zhuoyu</creatorcontrib><creatorcontrib>Wu, Qinchen</creatorcontrib><creatorcontrib>Fang, Wenqi</creatorcontrib><creatorcontrib>Ou, Wenhui</creatorcontrib><creatorcontrib>Wang, Quanjun</creatorcontrib><creatorcontrib>Zhang, Linde</creatorcontrib><creatorcontrib>Chen, Chao</creatorcontrib><creatorcontrib>Wang, Zheng</creatorcontrib><creatorcontrib>Li, Heshan</creatorcontrib><title>Harmonizing Unets: Attention Fusion module in cascaded-Unets for low-quality OCT image fluid segmentation</title><title>Computers in biology and medicine</title><addtitle>Comput Biol Med</addtitle><description>Optical coherence tomography (OCT) is widely used for its high resolution. Accurate OCT image segmentation can significantly improve the diagnosis and treatment of retinal diseases such as Diabetic Macular Edema (DME). However, in resource-limited regions, portable devices with low-quality output are more frequently used, severely affecting the performance of segmentation. To address this issue, we propose a novel methodology in this paper, including a dedicated pre-processing pipeline and an end-to-end double U-shaped cascaded architecture, H-Unets. In addition, an Adaptive Attention Fusion (AAF) module is elaborately designed to improve the segmentation performance of H-Unets. To demonstrate the effectiveness of our method, we conduct a bunch of ablation and comparative studies on three open-source datasets. The experimental results show the validity of the pre-processing pipeline and H-Unets, achieving the highest Dice score of 90.60%±0.87% among popular methods in a relatively small model size.</description><subject>Ablation</subject><subject>Accuracy</subject><subject>Algorithms</subject><subject>Artificial intelligence</subject><subject>Biology</subject><subject>Comparative studies</subject><subject>Computers</subject><subject>Datasets</subject><subject>Deep learning</subject><subject>Diabetes</subject><subject>Diabetes mellitus</subject><subject>Diabetic Retinopathy - diagnostic imaging</subject><subject>Edema</subject><subject>Efficiency</subject><subject>Fusion module</subject><subject>Humans</subject><subject>Image processing</subject><subject>Image Processing, Computer-Assisted - methods</subject><subject>Image quality</subject><subject>Image resolution</subject><subject>Image segmentation</subject><subject>Intraretinal fluid detection</subject><subject>Low income groups</subject><subject>Macular Edema - diagnostic imaging</subject><subject>Medical image segmentation</subject><subject>Medical imaging</subject><subject>Medicine</subject><subject>Modules</subject><subject>Optical Coherence Tomography</subject><subject>Portable equipment</subject><subject>Public health</subject><subject>Tomography</subject><subject>Tomography, Optical Coherence - methods</subject><issn>0010-4825</issn><issn>1879-0534</issn><issn>1879-0534</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2024</creationdate><recordtype>article</recordtype><sourceid>EIF</sourceid><recordid>eNqFkUFv1DAQhS0EotvCX0CWuHDJ4rGdxOFWVrRFqtRLe7Yce7LyKrG3dgIqvx6HbYXEhdNI9jdvZt4jhALbAoPm82Fr43TsfZzQbTnjsjx3nItXZAOq7SpWC_mabBgDVknF6zNynvOBMSaZYG_JmehEowTwDfE3Jk0x-F8-7OlDwDl_oZfzjGH2MdCrJa9lim4ZkfpArcnWOHTVH5QOMdEx_qweFzP6-Yne7e6pn8we6TAu3tGM-6lImVXsHXkzmDHj--d6QR6uvt3vbqrbu-vvu8vbykI5oXKopGLoTM1lz5u2t7YdQILEduiMqDmg7XvVMdkODeu6FjsloeVgbAMNV-KCfDrpHlN8XDDPevLZ4jiagHHJWgAIoUAKVtCP_6CHuKRQtisUL77VAtpCqRNlU8w54aCPqRyZnjQwvcahD_pvHHqNQ5_iKK0fngcs_fr30vjifwG-ngAsjvzwmHS2HoNF5xPaWbvo_z_lN6s9n9I</recordid><startdate>202412</startdate><enddate>202412</enddate><creator>Wu, Zhuoyu</creator><creator>Wu, Qinchen</creator><creator>Fang, Wenqi</creator><creator>Ou, Wenhui</creator><creator>Wang, Quanjun</creator><creator>Zhang, Linde</creator><creator>Chen, Chao</creator><creator>Wang, Zheng</creator><creator>Li, Heshan</creator><general>Elsevier Ltd</general><general>Elsevier Limited</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>8FD</scope><scope>FR3</scope><scope>JQ2</scope><scope>K9.</scope><scope>M7Z</scope><scope>NAPCQ</scope><scope>P64</scope><scope>7X8</scope><orcidid>https://orcid.org/0009-0009-6779-5841</orcidid><orcidid>https://orcid.org/0009-0002-1978-8021</orcidid><orcidid>https://orcid.org/0009-0002-5147-4858</orcidid><orcidid>https://orcid.org/0000-0001-6488-224X</orcidid><orcidid>https://orcid.org/0000-0003-3347-0511</orcidid><orcidid>https://orcid.org/0000-0002-9075-7626</orcidid></search><sort><creationdate>202412</creationdate><title>Harmonizing Unets: Attention Fusion module in cascaded-Unets for low-quality OCT image fluid segmentation</title><author>Wu, Zhuoyu ; Wu, Qinchen ; Fang, Wenqi ; Ou, Wenhui ; Wang, Quanjun ; Zhang, Linde ; Chen, Chao ; Wang, Zheng ; Li, Heshan</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c1922-de8480eda524b267bcc7f1414e7f9a3521ecbb89047f60997e9841721ac616283</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2024</creationdate><topic>Ablation</topic><topic>Accuracy</topic><topic>Algorithms</topic><topic>Artificial intelligence</topic><topic>Biology</topic><topic>Comparative studies</topic><topic>Computers</topic><topic>Datasets</topic><topic>Deep learning</topic><topic>Diabetes</topic><topic>Diabetes mellitus</topic><topic>Diabetic Retinopathy - diagnostic imaging</topic><topic>Edema</topic><topic>Efficiency</topic><topic>Fusion module</topic><topic>Humans</topic><topic>Image processing</topic><topic>Image Processing, Computer-Assisted - methods</topic><topic>Image quality</topic><topic>Image resolution</topic><topic>Image segmentation</topic><topic>Intraretinal fluid detection</topic><topic>Low income groups</topic><topic>Macular Edema - diagnostic imaging</topic><topic>Medical image segmentation</topic><topic>Medical imaging</topic><topic>Medicine</topic><topic>Modules</topic><topic>Optical Coherence Tomography</topic><topic>Portable equipment</topic><topic>Public health</topic><topic>Tomography</topic><topic>Tomography, Optical Coherence - methods</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Wu, Zhuoyu</creatorcontrib><creatorcontrib>Wu, Qinchen</creatorcontrib><creatorcontrib>Fang, Wenqi</creatorcontrib><creatorcontrib>Ou, Wenhui</creatorcontrib><creatorcontrib>Wang, Quanjun</creatorcontrib><creatorcontrib>Zhang, Linde</creatorcontrib><creatorcontrib>Chen, Chao</creatorcontrib><creatorcontrib>Wang, Zheng</creatorcontrib><creatorcontrib>Li, Heshan</creatorcontrib><collection>Medline</collection><collection>MEDLINE</collection><collection>MEDLINE (Ovid)</collection><collection>MEDLINE</collection><collection>MEDLINE</collection><collection>PubMed</collection><collection>CrossRef</collection><collection>Technology Research Database</collection><collection>Engineering Research Database</collection><collection>ProQuest Computer Science Collection</collection><collection>ProQuest Health & Medical Complete (Alumni)</collection><collection>Biochemistry Abstracts 1</collection><collection>Nursing & Allied Health Premium</collection><collection>Biotechnology and BioEngineering Abstracts</collection><collection>MEDLINE - Academic</collection><jtitle>Computers in biology and medicine</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Wu, Zhuoyu</au><au>Wu, Qinchen</au><au>Fang, Wenqi</au><au>Ou, Wenhui</au><au>Wang, Quanjun</au><au>Zhang, Linde</au><au>Chen, Chao</au><au>Wang, Zheng</au><au>Li, Heshan</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Harmonizing Unets: Attention Fusion module in cascaded-Unets for low-quality OCT image fluid segmentation</atitle><jtitle>Computers in biology and medicine</jtitle><addtitle>Comput Biol Med</addtitle><date>2024-12</date><risdate>2024</risdate><volume>183</volume><spage>109223</spage><pages>109223-</pages><artnum>109223</artnum><issn>0010-4825</issn><issn>1879-0534</issn><eissn>1879-0534</eissn><abstract>Optical coherence tomography (OCT) is widely used for its high resolution. Accurate OCT image segmentation can significantly improve the diagnosis and treatment of retinal diseases such as Diabetic Macular Edema (DME). However, in resource-limited regions, portable devices with low-quality output are more frequently used, severely affecting the performance of segmentation. To address this issue, we propose a novel methodology in this paper, including a dedicated pre-processing pipeline and an end-to-end double U-shaped cascaded architecture, H-Unets. In addition, an Adaptive Attention Fusion (AAF) module is elaborately designed to improve the segmentation performance of H-Unets. To demonstrate the effectiveness of our method, we conduct a bunch of ablation and comparative studies on three open-source datasets. The experimental results show the validity of the pre-processing pipeline and H-Unets, achieving the highest Dice score of 90.60%±0.87% among popular methods in a relatively small model size.</abstract><cop>United States</cop><pub>Elsevier Ltd</pub><pmid>39368312</pmid><doi>10.1016/j.compbiomed.2024.109223</doi><orcidid>https://orcid.org/0009-0009-6779-5841</orcidid><orcidid>https://orcid.org/0009-0002-1978-8021</orcidid><orcidid>https://orcid.org/0009-0002-5147-4858</orcidid><orcidid>https://orcid.org/0000-0001-6488-224X</orcidid><orcidid>https://orcid.org/0000-0003-3347-0511</orcidid><orcidid>https://orcid.org/0000-0002-9075-7626</orcidid></addata></record> |
| fulltext | fulltext |
| identifier | ISSN: 0010-4825 |
| ispartof | Computers in biology and medicine, 2024-12, Vol.183, p.109223, Article 109223 |
| issn | 0010-4825 1879-0534 1879-0534 |
| language | eng |
| recordid | cdi_proquest_miscellaneous_3113381430 |
| source | MEDLINE; ScienceDirect Journals (5 years ago - present) |
| subjects | Ablation Accuracy Algorithms Artificial intelligence Biology Comparative studies Computers Datasets Deep learning Diabetes Diabetes mellitus Diabetic Retinopathy - diagnostic imaging Edema Efficiency Fusion module Humans Image processing Image Processing, Computer-Assisted - methods Image quality Image resolution Image segmentation Intraretinal fluid detection Low income groups Macular Edema - diagnostic imaging Medical image segmentation Medical imaging Medicine Modules Optical Coherence Tomography Portable equipment Public health Tomography Tomography, Optical Coherence - methods |
| title | Harmonizing Unets: Attention Fusion module in cascaded-Unets for low-quality OCT image fluid segmentation |
| url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-08T01%3A02%3A13IST&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=Harmonizing%20Unets:%20Attention%20Fusion%20module%20in%20cascaded-Unets%20for%20low-quality%20OCT%20image%20fluid%20segmentation&rft.jtitle=Computers%20in%20biology%20and%20medicine&rft.au=Wu,%20Zhuoyu&rft.date=2024-12&rft.volume=183&rft.spage=109223&rft.pages=109223-&rft.artnum=109223&rft.issn=0010-4825&rft.eissn=1879-0534&rft_id=info:doi/10.1016/j.compbiomed.2024.109223&rft_dat=%3Cproquest_cross%3E3113381430%3C/proquest_cross%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=3128255317&rft_id=info:pmid/39368312&rft_els_id=S0010482524013088&rfr_iscdi=true |