Hybrid deep learning and optimal graph search method for optical coherence tomography layer segmentation in diseases affecting the optic nerve
Accurate segmentation of retinal layers in optical coherence tomography (OCT) images is critical for assessing diseases that affect the optic nerve, but existing automated algorithms often fail when pathology causes irregular layer topology, such as extreme thinning of the ganglion cell-inner plexif...
Gespeichert in:
Veröffentlicht in: | Biomedical optics express 2024-06, Vol.15 (6), p.3681-3698 |
---|---|
Hauptverfasser: | , , , , , , , , , |
Format: | Artikel |
Sprache: | eng |
Online-Zugang: | Volltext |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
container_end_page | 3698 |
---|---|
container_issue | 6 |
container_start_page | 3681 |
container_title | Biomedical optics express |
container_volume | 15 |
creator | Chen, Zhi Zhang, Honghai Linton, Edward F Johnson, Brett A Choi, Yun Jae Kupersmith, Mark J Sonka, Milan Garvin, Mona K Kardon, Randy H Wang, Jui-Kai |
description | Accurate segmentation of retinal layers in optical coherence tomography (OCT) images is critical for assessing diseases that affect the optic nerve, but existing automated algorithms often fail when pathology causes irregular layer topology, such as extreme thinning of the ganglion cell-inner plexiform layer (GCIPL). Deep LOGISMOS, a hybrid approach that combines the strengths of deep learning and 3D graph search to overcome their limitations, was developed to improve the accuracy, robustness and generalizability of retinal layer segmentation. The method was trained on 124 OCT volumes from both eyes of 31 non-arteritic anterior ischemic optic neuropathy (NAION) patients and tested on three cross-sectional datasets with available reference tracings: Test-NAION (40 volumes from both eyes of 20 NAION subjects), Test-G (29 volumes from 29 glaucoma subjects/eyes), and Test-JHU (35 volumes from 21 multiple sclerosis and 14 control subjects/eyes) and one longitudinal dataset without reference tracings: Test-G-L (155 volumes from 15 glaucoma patients/eyes). In the three test datasets with reference tracings (Test-NAION, Test-G, and Test-JHU), Deep LOGISMOS achieved very high Dice similarity coefficients (%) on GCIPL: 89.97±3.59, 90.63±2.56, and 94.06±1.76, respectively. In the same context, Deep LOGISMOS outperformed the Iowa reference algorithms by improving the Dice score by 17.5, 5.4, and 7.5, and also surpassed the deep learning framework nnU-Net with improvements of 4.4, 3.7, and 1.0. For the 15 severe glaucoma eyes with marked GCIPL thinning (Test-G-L), it demonstrated reliable regional GCIPL thickness measurement over five years. The proposed Deep LOGISMOS approach has potential to enhance precise quantification of retinal structures, aiding diagnosis and treatment management of optic nerve diseases. |
doi_str_mv | 10.1364/BOE.516045 |
format | Article |
fullrecord | <record><control><sourceid>proquest_pubme</sourceid><recordid>TN_cdi_pubmedcentral_primary_oai_pubmedcentral_nih_gov_11166436</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>3067917356</sourcerecordid><originalsourceid>FETCH-LOGICAL-c268t-7dcdd3cc37871d8a659f07fd8734a0e31cf95f61da3a88db5dbf91bbe49ec78a3</originalsourceid><addsrcrecordid>eNpVkcFqGzEQhkVJaUKaSx-g6BgCTqVoV9KeShLSuhDIpT0LrTTyquxKW0k2-CXyzJHt1CRz0cD_zT8jfoS-UHJNGW--3T09XLeUk6b9gM5uaMsXgsj25E1_ii5y_ktqNY0gTH5Cp0xKLmqdoefltk_eYgsw4xF0Cj6ssA4Wx7n4SY94lfQ84FwlM-AJyhAtdjHtdVN1EwdIEAzgEqe4p7d41FtIdWg1QSi6-BiwD9j6apMhY-0cmLLbVAY4OOEAaQOf0UenxwwXr-85-vPj4ff9cvH49PPX_e3jwtxwWRbCGmuZMUxIQa3UvO0cEc5KwRpNgFHjutZxajXTUtq-tb3raN9D04ERUrNz9P3gO6_7CaypVyY9qjnVL6etitqr90rwg1rFjaKUct4wXh0uXx1S_LeGXNTks4Fx1AHiOitGuOioYO0OvTqgJsWcE7jjHkrULkRVQ1SHECv89e1lR_R_ZOwFH2OcXg</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>3067917356</pqid></control><display><type>article</type><title>Hybrid deep learning and optimal graph search method for optical coherence tomography layer segmentation in diseases affecting the optic nerve</title><source>DOAJ Directory of Open Access Journals</source><source>Elektronische Zeitschriftenbibliothek - Frei zugängliche E-Journals</source><source>PubMed Central</source><creator>Chen, Zhi ; Zhang, Honghai ; Linton, Edward F ; Johnson, Brett A ; Choi, Yun Jae ; Kupersmith, Mark J ; Sonka, Milan ; Garvin, Mona K ; Kardon, Randy H ; Wang, Jui-Kai</creator><creatorcontrib>Chen, Zhi ; Zhang, Honghai ; Linton, Edward F ; Johnson, Brett A ; Choi, Yun Jae ; Kupersmith, Mark J ; Sonka, Milan ; Garvin, Mona K ; Kardon, Randy H ; Wang, Jui-Kai</creatorcontrib><description>Accurate segmentation of retinal layers in optical coherence tomography (OCT) images is critical for assessing diseases that affect the optic nerve, but existing automated algorithms often fail when pathology causes irregular layer topology, such as extreme thinning of the ganglion cell-inner plexiform layer (GCIPL). Deep LOGISMOS, a hybrid approach that combines the strengths of deep learning and 3D graph search to overcome their limitations, was developed to improve the accuracy, robustness and generalizability of retinal layer segmentation. The method was trained on 124 OCT volumes from both eyes of 31 non-arteritic anterior ischemic optic neuropathy (NAION) patients and tested on three cross-sectional datasets with available reference tracings: Test-NAION (40 volumes from both eyes of 20 NAION subjects), Test-G (29 volumes from 29 glaucoma subjects/eyes), and Test-JHU (35 volumes from 21 multiple sclerosis and 14 control subjects/eyes) and one longitudinal dataset without reference tracings: Test-G-L (155 volumes from 15 glaucoma patients/eyes). In the three test datasets with reference tracings (Test-NAION, Test-G, and Test-JHU), Deep LOGISMOS achieved very high Dice similarity coefficients (%) on GCIPL: 89.97±3.59, 90.63±2.56, and 94.06±1.76, respectively. In the same context, Deep LOGISMOS outperformed the Iowa reference algorithms by improving the Dice score by 17.5, 5.4, and 7.5, and also surpassed the deep learning framework nnU-Net with improvements of 4.4, 3.7, and 1.0. For the 15 severe glaucoma eyes with marked GCIPL thinning (Test-G-L), it demonstrated reliable regional GCIPL thickness measurement over five years. The proposed Deep LOGISMOS approach has potential to enhance precise quantification of retinal structures, aiding diagnosis and treatment management of optic nerve diseases.</description><identifier>ISSN: 2156-7085</identifier><identifier>EISSN: 2156-7085</identifier><identifier>DOI: 10.1364/BOE.516045</identifier><identifier>PMID: 38867777</identifier><language>eng</language><publisher>United States: Optica Publishing Group</publisher><ispartof>Biomedical optics express, 2024-06, Vol.15 (6), p.3681-3698</ispartof><rights>2024 Optica Publishing Group.</rights><rights>2024 Optica Publishing Group 2024 Optica Publishing Group</rights><lds50>peer_reviewed</lds50><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed><cites>FETCH-LOGICAL-c268t-7dcdd3cc37871d8a659f07fd8734a0e31cf95f61da3a88db5dbf91bbe49ec78a3</cites><orcidid>0000-0002-2725-6239</orcidid></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktopdf>$$Uhttps://www.ncbi.nlm.nih.gov/pmc/articles/PMC11166436/pdf/$$EPDF$$P50$$Gpubmedcentral$$Hfree_for_read</linktopdf><linktohtml>$$Uhttps://www.ncbi.nlm.nih.gov/pmc/articles/PMC11166436/$$EHTML$$P50$$Gpubmedcentral$$Hfree_for_read</linktohtml><link.rule.ids>230,314,723,776,780,860,881,27903,27904,53769,53771</link.rule.ids><backlink>$$Uhttps://www.ncbi.nlm.nih.gov/pubmed/38867777$$D View this record in MEDLINE/PubMed$$Hfree_for_read</backlink></links><search><creatorcontrib>Chen, Zhi</creatorcontrib><creatorcontrib>Zhang, Honghai</creatorcontrib><creatorcontrib>Linton, Edward F</creatorcontrib><creatorcontrib>Johnson, Brett A</creatorcontrib><creatorcontrib>Choi, Yun Jae</creatorcontrib><creatorcontrib>Kupersmith, Mark J</creatorcontrib><creatorcontrib>Sonka, Milan</creatorcontrib><creatorcontrib>Garvin, Mona K</creatorcontrib><creatorcontrib>Kardon, Randy H</creatorcontrib><creatorcontrib>Wang, Jui-Kai</creatorcontrib><title>Hybrid deep learning and optimal graph search method for optical coherence tomography layer segmentation in diseases affecting the optic nerve</title><title>Biomedical optics express</title><addtitle>Biomed Opt Express</addtitle><description>Accurate segmentation of retinal layers in optical coherence tomography (OCT) images is critical for assessing diseases that affect the optic nerve, but existing automated algorithms often fail when pathology causes irregular layer topology, such as extreme thinning of the ganglion cell-inner plexiform layer (GCIPL). Deep LOGISMOS, a hybrid approach that combines the strengths of deep learning and 3D graph search to overcome their limitations, was developed to improve the accuracy, robustness and generalizability of retinal layer segmentation. The method was trained on 124 OCT volumes from both eyes of 31 non-arteritic anterior ischemic optic neuropathy (NAION) patients and tested on three cross-sectional datasets with available reference tracings: Test-NAION (40 volumes from both eyes of 20 NAION subjects), Test-G (29 volumes from 29 glaucoma subjects/eyes), and Test-JHU (35 volumes from 21 multiple sclerosis and 14 control subjects/eyes) and one longitudinal dataset without reference tracings: Test-G-L (155 volumes from 15 glaucoma patients/eyes). In the three test datasets with reference tracings (Test-NAION, Test-G, and Test-JHU), Deep LOGISMOS achieved very high Dice similarity coefficients (%) on GCIPL: 89.97±3.59, 90.63±2.56, and 94.06±1.76, respectively. In the same context, Deep LOGISMOS outperformed the Iowa reference algorithms by improving the Dice score by 17.5, 5.4, and 7.5, and also surpassed the deep learning framework nnU-Net with improvements of 4.4, 3.7, and 1.0. For the 15 severe glaucoma eyes with marked GCIPL thinning (Test-G-L), it demonstrated reliable regional GCIPL thickness measurement over five years. The proposed Deep LOGISMOS approach has potential to enhance precise quantification of retinal structures, aiding diagnosis and treatment management of optic nerve diseases.</description><issn>2156-7085</issn><issn>2156-7085</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2024</creationdate><recordtype>article</recordtype><recordid>eNpVkcFqGzEQhkVJaUKaSx-g6BgCTqVoV9KeShLSuhDIpT0LrTTyquxKW0k2-CXyzJHt1CRz0cD_zT8jfoS-UHJNGW--3T09XLeUk6b9gM5uaMsXgsj25E1_ii5y_ktqNY0gTH5Cp0xKLmqdoefltk_eYgsw4xF0Cj6ssA4Wx7n4SY94lfQ84FwlM-AJyhAtdjHtdVN1EwdIEAzgEqe4p7d41FtIdWg1QSi6-BiwD9j6apMhY-0cmLLbVAY4OOEAaQOf0UenxwwXr-85-vPj4ff9cvH49PPX_e3jwtxwWRbCGmuZMUxIQa3UvO0cEc5KwRpNgFHjutZxajXTUtq-tb3raN9D04ERUrNz9P3gO6_7CaypVyY9qjnVL6etitqr90rwg1rFjaKUct4wXh0uXx1S_LeGXNTks4Fx1AHiOitGuOioYO0OvTqgJsWcE7jjHkrULkRVQ1SHECv89e1lR_R_ZOwFH2OcXg</recordid><startdate>20240601</startdate><enddate>20240601</enddate><creator>Chen, Zhi</creator><creator>Zhang, Honghai</creator><creator>Linton, Edward F</creator><creator>Johnson, Brett A</creator><creator>Choi, Yun Jae</creator><creator>Kupersmith, Mark J</creator><creator>Sonka, Milan</creator><creator>Garvin, Mona K</creator><creator>Kardon, Randy H</creator><creator>Wang, Jui-Kai</creator><general>Optica Publishing Group</general><scope>NPM</scope><scope>AAYXX</scope><scope>CITATION</scope><scope>7X8</scope><scope>5PM</scope><orcidid>https://orcid.org/0000-0002-2725-6239</orcidid></search><sort><creationdate>20240601</creationdate><title>Hybrid deep learning and optimal graph search method for optical coherence tomography layer segmentation in diseases affecting the optic nerve</title><author>Chen, Zhi ; Zhang, Honghai ; Linton, Edward F ; Johnson, Brett A ; Choi, Yun Jae ; Kupersmith, Mark J ; Sonka, Milan ; Garvin, Mona K ; Kardon, Randy H ; Wang, Jui-Kai</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c268t-7dcdd3cc37871d8a659f07fd8734a0e31cf95f61da3a88db5dbf91bbe49ec78a3</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2024</creationdate><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Chen, Zhi</creatorcontrib><creatorcontrib>Zhang, Honghai</creatorcontrib><creatorcontrib>Linton, Edward F</creatorcontrib><creatorcontrib>Johnson, Brett A</creatorcontrib><creatorcontrib>Choi, Yun Jae</creatorcontrib><creatorcontrib>Kupersmith, Mark J</creatorcontrib><creatorcontrib>Sonka, Milan</creatorcontrib><creatorcontrib>Garvin, Mona K</creatorcontrib><creatorcontrib>Kardon, Randy H</creatorcontrib><creatorcontrib>Wang, Jui-Kai</creatorcontrib><collection>PubMed</collection><collection>CrossRef</collection><collection>MEDLINE - Academic</collection><collection>PubMed Central (Full Participant titles)</collection><jtitle>Biomedical optics express</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Chen, Zhi</au><au>Zhang, Honghai</au><au>Linton, Edward F</au><au>Johnson, Brett A</au><au>Choi, Yun Jae</au><au>Kupersmith, Mark J</au><au>Sonka, Milan</au><au>Garvin, Mona K</au><au>Kardon, Randy H</au><au>Wang, Jui-Kai</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Hybrid deep learning and optimal graph search method for optical coherence tomography layer segmentation in diseases affecting the optic nerve</atitle><jtitle>Biomedical optics express</jtitle><addtitle>Biomed Opt Express</addtitle><date>2024-06-01</date><risdate>2024</risdate><volume>15</volume><issue>6</issue><spage>3681</spage><epage>3698</epage><pages>3681-3698</pages><issn>2156-7085</issn><eissn>2156-7085</eissn><abstract>Accurate segmentation of retinal layers in optical coherence tomography (OCT) images is critical for assessing diseases that affect the optic nerve, but existing automated algorithms often fail when pathology causes irregular layer topology, such as extreme thinning of the ganglion cell-inner plexiform layer (GCIPL). Deep LOGISMOS, a hybrid approach that combines the strengths of deep learning and 3D graph search to overcome their limitations, was developed to improve the accuracy, robustness and generalizability of retinal layer segmentation. The method was trained on 124 OCT volumes from both eyes of 31 non-arteritic anterior ischemic optic neuropathy (NAION) patients and tested on three cross-sectional datasets with available reference tracings: Test-NAION (40 volumes from both eyes of 20 NAION subjects), Test-G (29 volumes from 29 glaucoma subjects/eyes), and Test-JHU (35 volumes from 21 multiple sclerosis and 14 control subjects/eyes) and one longitudinal dataset without reference tracings: Test-G-L (155 volumes from 15 glaucoma patients/eyes). In the three test datasets with reference tracings (Test-NAION, Test-G, and Test-JHU), Deep LOGISMOS achieved very high Dice similarity coefficients (%) on GCIPL: 89.97±3.59, 90.63±2.56, and 94.06±1.76, respectively. In the same context, Deep LOGISMOS outperformed the Iowa reference algorithms by improving the Dice score by 17.5, 5.4, and 7.5, and also surpassed the deep learning framework nnU-Net with improvements of 4.4, 3.7, and 1.0. For the 15 severe glaucoma eyes with marked GCIPL thinning (Test-G-L), it demonstrated reliable regional GCIPL thickness measurement over five years. The proposed Deep LOGISMOS approach has potential to enhance precise quantification of retinal structures, aiding diagnosis and treatment management of optic nerve diseases.</abstract><cop>United States</cop><pub>Optica Publishing Group</pub><pmid>38867777</pmid><doi>10.1364/BOE.516045</doi><tpages>18</tpages><orcidid>https://orcid.org/0000-0002-2725-6239</orcidid><oa>free_for_read</oa></addata></record> |
fulltext | fulltext |
identifier | ISSN: 2156-7085 |
ispartof | Biomedical optics express, 2024-06, Vol.15 (6), p.3681-3698 |
issn | 2156-7085 2156-7085 |
language | eng |
recordid | cdi_pubmedcentral_primary_oai_pubmedcentral_nih_gov_11166436 |
source | DOAJ Directory of Open Access Journals; Elektronische Zeitschriftenbibliothek - Frei zugängliche E-Journals; PubMed Central |
title | Hybrid deep learning and optimal graph search method for optical coherence tomography layer segmentation in diseases affecting the optic nerve |
url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-28T04%3A28%3A58IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_pubme&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Hybrid%20deep%20learning%20and%20optimal%20graph%20search%20method%20for%20optical%20coherence%20tomography%20layer%20segmentation%20in%20diseases%20affecting%20the%20optic%20nerve&rft.jtitle=Biomedical%20optics%20express&rft.au=Chen,%20Zhi&rft.date=2024-06-01&rft.volume=15&rft.issue=6&rft.spage=3681&rft.epage=3698&rft.pages=3681-3698&rft.issn=2156-7085&rft.eissn=2156-7085&rft_id=info:doi/10.1364/BOE.516045&rft_dat=%3Cproquest_pubme%3E3067917356%3C/proquest_pubme%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=3067917356&rft_id=info:pmid/38867777&rfr_iscdi=true |