Quantifying the spatial patterns of retinal ganglion cell loss and progression in optic neuropathy by applying a deep learning variational autoencoder approach to optical coherence tomography

Glaucoma, optic neuritis (ON), and non-arteritic anterior ischemic optic neuropathy (NAION) produce distinct patterns of retinal ganglion cell (RGC) damage. We propose a booster Variational Autoencoder (bVAE) to capture spatial variations in RGC loss and generate latent space (LS) montage maps that...

Full description

Saved in:
Bibliographic Details
Published in:Frontiers in ophthalmology Vol. 4; p. 1497848
Main Authors: Wang, Jui-Kai, Johnson, Brett A., Chen, Zhi, Zhang, Honghai, Szanto, David, Woods, Brian, Wall, Michael, Kwon, Young H., Linton, Edward F., Pouw, Andrew, Kupersmith, Mark J., Garvin, Mona K., Kardon, Randy H.
Format: Journal Article
Language:English
Published: Switzerland Frontiers Media S.A 03.02.2025
Subjects:
ISSN:2674-0826, 2674-0826
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Abstract Glaucoma, optic neuritis (ON), and non-arteritic anterior ischemic optic neuropathy (NAION) produce distinct patterns of retinal ganglion cell (RGC) damage. We propose a booster Variational Autoencoder (bVAE) to capture spatial variations in RGC loss and generate latent space (LS) montage maps that visualize different degrees and spatial patterns of optic nerve bundle injury. Furthermore, the bVAE model is capable of tracking the spatial pattern of RGC thinning over time and classifying the underlying cause. The bVAE model consists of an encoder, a display decoder, and a booster decoder. The encoder decomposes input ganglion cell layer (GCL) thickness maps into two display latent variables (dLVs) and eight booster latent variables (bLVs). The dLVs capture primary spatial patterns of RGC thinning, while the display decoder reconstructs the GCL map and creates the LS montage map. The bLVs add finer spatial details, improving reconstruction accuracy. XGBoost was used to analyze the dLVs and bLVs, estimating normal/abnormal GCL thinning and classifying diseases (glaucoma, ON, and NAION). A total of 10,701 OCT macular scans from 822 subjects were included in this study. Incorporating bLVs improved reconstruction accuracy, with the image-based root-mean-square error (RMSE) between input and reconstructed GCL thickness maps decreasing from 5.55 ± 2.29 µm (two dLVs only) to 4.02 ± 1.61 µm (two dLVs and eight bLVs). However, the image-based structural similarity index (SSIM) remained similar (0.91 ± 0.04), indicating that just two dLVs effectively capture the main GCL spatial patterns. For classification, the XGBoost model achieved an AUC of 0.98 for identifying abnormal spatial patterns of GCL thinning over time using the dLVs. Disease classification yielded AUCs of 0.95 for glaucoma, 0.84 for ON, and 0.93 for NAION, with bLVs further increasing the AUCs to 0.96 for glaucoma, 0.93 for ON, and 0.99 for NAION. This study presents a novel approach to visualizing and quantifying GCL thinning patterns in optic neuropathies using the bVAE model. The combination of dLVs and bLVs enhances the model's ability to capture key spatial features and predict disease progression. Future work will focus on integrating additional image modalities to further refine the model's diagnostic capabilities.
AbstractList IntroductionGlaucoma, optic neuritis (ON), and non-arteritic anterior ischemic optic neuropathy (NAION) produce distinct patterns of retinal ganglion cell (RGC) damage. We propose a booster Variational Autoencoder (bVAE) to capture spatial variations in RGC loss and generate latent space (LS) montage maps that visualize different degrees and spatial patterns of optic nerve bundle injury. Furthermore, the bVAE model is capable of tracking the spatial pattern of RGC thinning over time and classifying the underlying cause.MethodsThe bVAE model consists of an encoder, a display decoder, and a booster decoder. The encoder decomposes input ganglion cell layer (GCL) thickness maps into two display latent variables (dLVs) and eight booster latent variables (bLVs). The dLVs capture primary spatial patterns of RGC thinning, while the display decoder reconstructs the GCL map and creates the LS montage map. The bLVs add finer spatial details, improving reconstruction accuracy. XGBoost was used to analyze the dLVs and bLVs, estimating normal/abnormal GCL thinning and classifying diseases (glaucoma, ON, and NAION). A total of 10,701 OCT macular scans from 822 subjects were included in this study.ResultsIncorporating bLVs improved reconstruction accuracy, with the image-based root-mean-square error (RMSE) between input and reconstructed GCL thickness maps decreasing from 5.55 ± 2.29 µm (two dLVs only) to 4.02 ± 1.61 µm (two dLVs and eight bLVs). However, the image-based structural similarity index (SSIM) remained similar (0.91 ± 0.04), indicating that just two dLVs effectively capture the main GCL spatial patterns. For classification, the XGBoost model achieved an AUC of 0.98 for identifying abnormal spatial patterns of GCL thinning over time using the dLVs. Disease classification yielded AUCs of 0.95 for glaucoma, 0.84 for ON, and 0.93 for NAION, with bLVs further increasing the AUCs to 0.96 for glaucoma, 0.93 for ON, and 0.99 for NAION.ConclusionThis study presents a novel approach to visualizing and quantifying GCL thinning patterns in optic neuropathies using the bVAE model. The combination of dLVs and bLVs enhances the model’s ability to capture key spatial features and predict disease progression. Future work will focus on integrating additional image modalities to further refine the model’s diagnostic capabilities.
Glaucoma, optic neuritis (ON), and non-arteritic anterior ischemic optic neuropathy (NAION) produce distinct patterns of retinal ganglion cell (RGC) damage. We propose a booster Variational Autoencoder (bVAE) to capture spatial variations in RGC loss and generate latent space (LS) montage maps that visualize different degrees and spatial patterns of optic nerve bundle injury. Furthermore, the bVAE model is capable of tracking the spatial pattern of RGC thinning over time and classifying the underlying cause. The bVAE model consists of an encoder, a display decoder, and a booster decoder. The encoder decomposes input ganglion cell layer (GCL) thickness maps into two display latent variables (dLVs) and eight booster latent variables (bLVs). The dLVs capture primary spatial patterns of RGC thinning, while the display decoder reconstructs the GCL map and creates the LS montage map. The bLVs add finer spatial details, improving reconstruction accuracy. XGBoost was used to analyze the dLVs and bLVs, estimating normal/abnormal GCL thinning and classifying diseases (glaucoma, ON, and NAION). A total of 10,701 OCT macular scans from 822 subjects were included in this study. Incorporating bLVs improved reconstruction accuracy, with the image-based root-mean-square error (RMSE) between input and reconstructed GCL thickness maps decreasing from 5.55 ± 2.29 µm (two dLVs only) to 4.02 ± 1.61 µm (two dLVs and eight bLVs). However, the image-based structural similarity index (SSIM) remained similar (0.91 ± 0.04), indicating that just two dLVs effectively capture the main GCL spatial patterns. For classification, the XGBoost model achieved an AUC of 0.98 for identifying abnormal spatial patterns of GCL thinning over time using the dLVs. Disease classification yielded AUCs of 0.95 for glaucoma, 0.84 for ON, and 0.93 for NAION, with bLVs further increasing the AUCs to 0.96 for glaucoma, 0.93 for ON, and 0.99 for NAION. This study presents a novel approach to visualizing and quantifying GCL thinning patterns in optic neuropathies using the bVAE model. The combination of dLVs and bLVs enhances the model's ability to capture key spatial features and predict disease progression. Future work will focus on integrating additional image modalities to further refine the model's diagnostic capabilities.
Glaucoma, optic neuritis (ON), and non-arteritic anterior ischemic optic neuropathy (NAION) produce distinct patterns of retinal ganglion cell (RGC) damage. We propose a booster Variational Autoencoder (bVAE) to capture spatial variations in RGC loss and generate latent space (LS) montage maps that visualize different degrees and spatial patterns of optic nerve bundle injury. Furthermore, the bVAE model is capable of tracking the spatial pattern of RGC thinning over time and classifying the underlying cause.IntroductionGlaucoma, optic neuritis (ON), and non-arteritic anterior ischemic optic neuropathy (NAION) produce distinct patterns of retinal ganglion cell (RGC) damage. We propose a booster Variational Autoencoder (bVAE) to capture spatial variations in RGC loss and generate latent space (LS) montage maps that visualize different degrees and spatial patterns of optic nerve bundle injury. Furthermore, the bVAE model is capable of tracking the spatial pattern of RGC thinning over time and classifying the underlying cause.The bVAE model consists of an encoder, a display decoder, and a booster decoder. The encoder decomposes input ganglion cell layer (GCL) thickness maps into two display latent variables (dLVs) and eight booster latent variables (bLVs). The dLVs capture primary spatial patterns of RGC thinning, while the display decoder reconstructs the GCL map and creates the LS montage map. The bLVs add finer spatial details, improving reconstruction accuracy. XGBoost was used to analyze the dLVs and bLVs, estimating normal/abnormal GCL thinning and classifying diseases (glaucoma, ON, and NAION). A total of 10,701 OCT macular scans from 822 subjects were included in this study.MethodsThe bVAE model consists of an encoder, a display decoder, and a booster decoder. The encoder decomposes input ganglion cell layer (GCL) thickness maps into two display latent variables (dLVs) and eight booster latent variables (bLVs). The dLVs capture primary spatial patterns of RGC thinning, while the display decoder reconstructs the GCL map and creates the LS montage map. The bLVs add finer spatial details, improving reconstruction accuracy. XGBoost was used to analyze the dLVs and bLVs, estimating normal/abnormal GCL thinning and classifying diseases (glaucoma, ON, and NAION). A total of 10,701 OCT macular scans from 822 subjects were included in this study.Incorporating bLVs improved reconstruction accuracy, with the image-based root-mean-square error (RMSE) between input and reconstructed GCL thickness maps decreasing from 5.55 ± 2.29 µm (two dLVs only) to 4.02 ± 1.61 µm (two dLVs and eight bLVs). However, the image-based structural similarity index (SSIM) remained similar (0.91 ± 0.04), indicating that just two dLVs effectively capture the main GCL spatial patterns. For classification, the XGBoost model achieved an AUC of 0.98 for identifying abnormal spatial patterns of GCL thinning over time using the dLVs. Disease classification yielded AUCs of 0.95 for glaucoma, 0.84 for ON, and 0.93 for NAION, with bLVs further increasing the AUCs to 0.96 for glaucoma, 0.93 for ON, and 0.99 for NAION.ResultsIncorporating bLVs improved reconstruction accuracy, with the image-based root-mean-square error (RMSE) between input and reconstructed GCL thickness maps decreasing from 5.55 ± 2.29 µm (two dLVs only) to 4.02 ± 1.61 µm (two dLVs and eight bLVs). However, the image-based structural similarity index (SSIM) remained similar (0.91 ± 0.04), indicating that just two dLVs effectively capture the main GCL spatial patterns. For classification, the XGBoost model achieved an AUC of 0.98 for identifying abnormal spatial patterns of GCL thinning over time using the dLVs. Disease classification yielded AUCs of 0.95 for glaucoma, 0.84 for ON, and 0.93 for NAION, with bLVs further increasing the AUCs to 0.96 for glaucoma, 0.93 for ON, and 0.99 for NAION.This study presents a novel approach to visualizing and quantifying GCL thinning patterns in optic neuropathies using the bVAE model. The combination of dLVs and bLVs enhances the model's ability to capture key spatial features and predict disease progression. Future work will focus on integrating additional image modalities to further refine the model's diagnostic capabilities.ConclusionThis study presents a novel approach to visualizing and quantifying GCL thinning patterns in optic neuropathies using the bVAE model. The combination of dLVs and bLVs enhances the model's ability to capture key spatial features and predict disease progression. Future work will focus on integrating additional image modalities to further refine the model's diagnostic capabilities.
Author Woods, Brian
Linton, Edward F.
Szanto, David
Wall, Michael
Wang, Jui-Kai
Johnson, Brett A.
Chen, Zhi
Pouw, Andrew
Zhang, Honghai
Kwon, Young H.
Kardon, Randy H.
Kupersmith, Mark J.
Garvin, Mona K.
AuthorAffiliation 5 Department of Neurology, Icahn School of Medicine at Mount Sinai , New York, NY , United States
3 Department of Electrical and Computer Engineering, University of Iowa , Iowa City, IA , United States
9 Department of Neurosurgery, Icahn School of Medicine at Mount Sinai , New York, NY , United States
4 Iowa Institute for Biomedical Imaging, University of Iowa , Iowa City, IA , United States
1 Center for the Prevention and Treatment of Visual Loss, Iowa City VA Health Care System , Iowa City, IA , United States
6 Department of Ophthalmology, University Hospital Galway , Galway , Ireland
7 Department of Physics, School of Natural Sciences, University of Galway , Galway , Ireland
2 Department of Ophthalmology and Visual Sciences, University of Iowa , Iowa City, IA , United States
8 Department of Ophthalmology, Icahn School of Medicine at Mount Sinai , New York, NY , United States
AuthorAffiliation_xml – name: 6 Department of Ophthalmology, University Hospital Galway , Galway , Ireland
– name: 8 Department of Ophthalmology, Icahn School of Medicine at Mount Sinai , New York, NY , United States
– name: 4 Iowa Institute for Biomedical Imaging, University of Iowa , Iowa City, IA , United States
– name: 2 Department of Ophthalmology and Visual Sciences, University of Iowa , Iowa City, IA , United States
– name: 3 Department of Electrical and Computer Engineering, University of Iowa , Iowa City, IA , United States
– name: 7 Department of Physics, School of Natural Sciences, University of Galway , Galway , Ireland
– name: 1 Center for the Prevention and Treatment of Visual Loss, Iowa City VA Health Care System , Iowa City, IA , United States
– name: 5 Department of Neurology, Icahn School of Medicine at Mount Sinai , New York, NY , United States
– name: 9 Department of Neurosurgery, Icahn School of Medicine at Mount Sinai , New York, NY , United States
Author_xml – sequence: 1
  givenname: Jui-Kai
  surname: Wang
  fullname: Wang, Jui-Kai
– sequence: 2
  givenname: Brett A.
  surname: Johnson
  fullname: Johnson, Brett A.
– sequence: 3
  givenname: Zhi
  surname: Chen
  fullname: Chen, Zhi
– sequence: 4
  givenname: Honghai
  surname: Zhang
  fullname: Zhang, Honghai
– sequence: 5
  givenname: David
  surname: Szanto
  fullname: Szanto, David
– sequence: 6
  givenname: Brian
  surname: Woods
  fullname: Woods, Brian
– sequence: 7
  givenname: Michael
  surname: Wall
  fullname: Wall, Michael
– sequence: 8
  givenname: Young H.
  surname: Kwon
  fullname: Kwon, Young H.
– sequence: 9
  givenname: Edward F.
  surname: Linton
  fullname: Linton, Edward F.
– sequence: 10
  givenname: Andrew
  surname: Pouw
  fullname: Pouw, Andrew
– sequence: 11
  givenname: Mark J.
  surname: Kupersmith
  fullname: Kupersmith, Mark J.
– sequence: 12
  givenname: Mona K.
  surname: Garvin
  fullname: Garvin, Mona K.
– sequence: 13
  givenname: Randy H.
  surname: Kardon
  fullname: Kardon, Randy H.
BackLink https://www.ncbi.nlm.nih.gov/pubmed/39963427$$D View this record in MEDLINE/PubMed
BookMark eNp9ks9u1DAQxiNUREvpC3BAPnLZxbG9iX1CqOJPpUoICc7WxJkkrrJ2sJ1Kebq-Gs7uUrUcODmZ-eb3OZnvdXHmvMOieFvSLedSfej8NKQto0xsS6FqKeSL4oJVtdhQyaqzJ8_nxVWMd5RSJuudEupVcc6Vqrhg9UXx8GMGl2y3WNeTNCCJEyQLI8lHwuAi8R0JmKzLtR5cP1rviMFxJKOPkYBryRR8HzDGtWMd8VOyhjicg8-QYSHNQmCaxoMFkBZxIiNCcOv7PQSbDf2Khzl5dMa3GNaB4MEMJPkjMPeNHzBkAebiPlvCNCxvipcdjBGvTudl8evL55_X3za337_eXH-63RhRqbRhHIUCI6AGKCXSKv8BWdasUtwIirysKlXXnArOBCvRSJQNVDvY0a4puUB-Wdwcua2HOz0Fu4ewaA9WHwo-9BpCvuaImoOiDWuwhc6IjFOUl1Sx7ATYoqGZ9fHImuZmj61BlwKMz6DPO84Ouvf3uiwlp7XgmfD-RAj-94wx6b2N61LAoZ-jzt8jKasYk1n67qnZo8vfCGQBOwpMyAsN2D1KSqrXqOlD1PQaNX2KWh6S_wwZmw57zBe24_9G_wDm_OBU
CitedBy_id crossref_primary_10_1001_jamaophthalmol_2025_2766
Cites_doi 10.1038/s41598-020-66355-5
10.1109/TIP.2003.819861
10.1016/j.preteyeres.2012.08.003
10.1364/BOE.487518
10.1038/s41598-019-54653-6
10.1167/tvst.11.8.30
10.1109/MSP.2008.930649
10.1097/IJG.0000000000002348
10.1167/iovs.15-18736
10.21037/aes.2018.05.08
10.1167/iovs.11-7962
10.1609/aaai.v32i1.11867
10.1177/1352458515598020
10.1167/tvst.13.1.13
10.1097/WNO.0b013e318238b9cb
10.1561/2200000056
10.3390/jimaging8050139
10.1167/iovs.16-21089
10.1364/BOE.516045
10.1145/3292500.3330757
10.1109/IJCNN48605.2020.9206694
10.1145/2939672.2939785
10.1016/j.ogla.2020.01.001
10.1109/CVPR46437.2021.00437
10.2147/OPTH.S42522
10.21105/joss.03021
10.1038/s41598-023-46903-5
10.1007/978-3-030-87000-3_17
10.1016/j.ophtha.2024.01.011
10.1016/j.xops.2023.100423
ContentType Journal Article
Copyright Copyright © 2025 Wang, Johnson, Chen, Zhang, Szanto, Woods, Wall, Kwon, Linton, Pouw, Kupersmith, Garvin and Kardon.
Copyright © 2025 Wang, Johnson, Chen, Zhang, Szanto, Woods, Wall, Kwon, Linton, Pouw, Kupersmith, Garvin and Kardon 2025 Wang, Johnson, Chen, Zhang, Szanto, Woods, Wall, Kwon, Linton, Pouw, Kupersmith, Garvin and Kardon
Copyright_xml – notice: Copyright © 2025 Wang, Johnson, Chen, Zhang, Szanto, Woods, Wall, Kwon, Linton, Pouw, Kupersmith, Garvin and Kardon.
– notice: Copyright © 2025 Wang, Johnson, Chen, Zhang, Szanto, Woods, Wall, Kwon, Linton, Pouw, Kupersmith, Garvin and Kardon 2025 Wang, Johnson, Chen, Zhang, Szanto, Woods, Wall, Kwon, Linton, Pouw, Kupersmith, Garvin and Kardon
DBID AAYXX
CITATION
NPM
7X8
5PM
DOA
DOI 10.3389/fopht.2024.1497848
DatabaseName CrossRef
PubMed
MEDLINE - Academic
PubMed Central (Full Participant titles)
DOAJ Directory of Open Access Journals
DatabaseTitle CrossRef
PubMed
MEDLINE - Academic
DatabaseTitleList
PubMed
MEDLINE - Academic
Database_xml – sequence: 1
  dbid: DOA
  name: DOAJ Directory of Open Access Journals
  url: https://www.doaj.org/
  sourceTypes: Open Website
– sequence: 2
  dbid: NPM
  name: PubMed
  url: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed
  sourceTypes: Index Database
– sequence: 3
  dbid: 7X8
  name: MEDLINE - Academic
  url: https://search.proquest.com/medline
  sourceTypes: Aggregation Database
DeliveryMethod fulltext_linktorsrc
Discipline Medicine
EISSN 2674-0826
ExternalDocumentID oai_doaj_org_article_3a90b2bedafc4242903109293caedec0
PMC11830743
39963427
10_3389_fopht_2024_1497848
Genre Journal Article
GrantInformation_xml – fundername: NEI NIH HHS
  grantid: R21 EY032522
– fundername: RRD VA
  grantid: IK1 RX005029
– fundername: NEI NIH HHS
  grantid: R01 EY031544
– fundername: NEI NIH HHS
  grantid: R01 EY023279
– fundername: RRD VA
  grantid: I01 RX003797
– fundername: RRD VA
  grantid: I01 RX001786
– fundername: RRD VA
  grantid: I50 RX003002
GroupedDBID 9T4
AAFWJ
AAYXX
AFPKN
ALMA_UNASSIGNED_HOLDINGS
CITATION
GROUPED_DOAJ
PGMZT
RPM
NPM
7X8
5PM
ID FETCH-LOGICAL-c469t-23e49ac4a7aa18e063998172693c40e316697730432421ec8e8ba65a50fb134e3
IEDL.DBID DOA
ISICitedReferencesCount 1
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=001422336000001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN 2674-0826
IngestDate Fri Oct 03 12:45:51 EDT 2025
Thu Aug 21 18:28:45 EDT 2025
Thu Oct 02 10:32:57 EDT 2025
Mon Jul 21 05:51:00 EDT 2025
Sat Nov 29 06:35:08 EST 2025
Tue Nov 18 22:41:46 EST 2025
IsDoiOpenAccess true
IsOpenAccess true
IsPeerReviewed true
IsScholarly true
Keywords optic neuritis (ON)
retinal ganglion cell (RGC) loss
non-arteritic anterior ischemic optic neuropathy (NAION)
variational autoencoder (VAE)
glaucoma
optical coherence tomography (OCT)
Language English
License Copyright © 2025 Wang, Johnson, Chen, Zhang, Szanto, Woods, Wall, Kwon, Linton, Pouw, Kupersmith, Garvin and Kardon.
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-c469t-23e49ac4a7aa18e063998172693c40e316697730432421ec8e8ba65a50fb134e3
Notes ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
Edited by: Yukihiro Shiga, University of Montreal Hospital Research Centre (CRCHUM), Canada
These authors have contributed equally to this work and share senior/last authorship
Fabio Lavinsky, University of the Rio dos Sinos Valley, Brazil
Reviewed by: Takashi Nishida, University of California, San Diego, United States
OpenAccessLink https://doaj.org/article/3a90b2bedafc4242903109293caedec0
PMID 39963427
PQID 3168026228
PQPubID 23479
ParticipantIDs doaj_primary_oai_doaj_org_article_3a90b2bedafc4242903109293caedec0
pubmedcentral_primary_oai_pubmedcentral_nih_gov_11830743
proquest_miscellaneous_3168026228
pubmed_primary_39963427
crossref_primary_10_3389_fopht_2024_1497848
crossref_citationtrail_10_3389_fopht_2024_1497848
PublicationCentury 2000
PublicationDate 2025-02-03
PublicationDateYYYYMMDD 2025-02-03
PublicationDate_xml – month: 02
  year: 2025
  text: 2025-02-03
  day: 03
PublicationDecade 2020
PublicationPlace Switzerland
PublicationPlace_xml – name: Switzerland
PublicationTitle Frontiers in ophthalmology
PublicationTitleAlternate Front Ophthalmol (Lausanne)
PublicationYear 2025
Publisher Frontiers Media S.A
Publisher_xml – name: Frontiers Media S.A
References Wang (B36) 2015
Daniel (B35) 2021
Hood (B2) 2013; 32
Burgess (B16) 2018
Wang (B37) 2017; 58
Mandal (B27) 2021; 1
Shon (B21) 2022; 11
Wy (B7) 2024; 33
Wang (B34) 2009; 26
Wang (B33) 2004; 13
Hallett (B20) 2020
Mishra (B11) 2020; 10
Asaoka (B26) 2020; 3
Hou (B17) 2016
Berchuck (B23) 2019
Wang (B6) 2021
B32
Yadav (B12) 2022; 8
Chen (B5) 2018; 3
Bouchacourt (B18) 2018
Chen (B10) 2024; 15
Wang (B29) 2024; 13
Kingma (B15) 2019; 12
Kardon (B1) 2011; 31
Dotan (B9) 2013; 7
He (B13) 2023; 14
Schorr (B8) 2022; 30
Mwanza (B14) 2011; 52
Huang (B19) 2018
Agharezaei (B22) 2023; 13
Zheng (B25) 2019
Kupersmith (B31) 2024; 131
Chen (B30) 2016
Kupersmith (B3) 2016; 57
Mohammadzadeh (B28) 2024; 4
Odaibo (B24) 2019; 1907
Kupersmith (B4) 2016; 22
References_xml – volume: 10
  start-page: 9541
  year: 2020
  ident: B11
  article-title: Automated retinal layer segmentation using graph-based algorithm incorporating deep-learning-derived information
  publication-title: Sci Rep
  doi: 10.1038/s41598-020-66355-5
– volume: 13
  year: 2004
  ident: B33
  article-title: Image quality assessment: From error visibility to structural similarity
  publication-title: IEEE Trans Image Process
  doi: 10.1109/TIP.2003.819861
– volume: 32
  start-page: 1
  year: 2013
  ident: B2
  article-title: Glaucomatous damage of the macula
  publication-title: Prog Retin Eye Res
  doi: 10.1016/j.preteyeres.2012.08.003
– volume: 14
  year: 2023
  ident: B13
  article-title: Longitudinal deep network for consistent OCT layer segmentation
  publication-title: BioMed Opt Express
  doi: 10.1364/BOE.487518
– start-page: 1
  year: 2019
  ident: B23
  article-title: Estimating rates of progression and predicting future visual fields in glaucoma using a deep variational autoencoder
  publication-title: Sci Rep
  doi: 10.1038/s41598-019-54653-6
– volume: 11
  year: 2022
  ident: B21
  article-title: Development of cumulative order-preserving image transformation based variational autoencoder for anterior segment optical coherence tomography images
  publication-title: Trans Vision Sci Technol
  doi: 10.1167/tvst.11.8.30
– start-page: 1
  year: 2018
  ident: B19
  article-title: Introvae: Introspective variational autoencoders for photographic image synthesis
  publication-title: Neural Inf Process Syst (NeurIPS)
– volume: 26
  start-page: 98
  year: 2009
  ident: B34
  article-title: Mean squared error: Love it or leave it? A new look at signal fidelity measures
  publication-title: IEEE Signal Process Magazine
  doi: 10.1109/MSP.2008.930649
– volume: 33
  year: 2024
  ident: B7
  article-title: Comparison of patterns of structural progression in primary open angle glaucoma and pseudoexfoliation glaucoma
  publication-title: J Glaucoma
  doi: 10.1097/IJG.0000000000002348
– volume: 1
  year: 2021
  ident: B27
  article-title: Assessing glaucoma in retinal fundus photographs using deep feature consistent variational autoencoders
  publication-title: ArXiv
– volume: 57
  year: 2016
  ident: B3
  article-title: Retinal ganglion cell layer thinning within one month of presentation for non-arteritic anterior ischemic optic neuropathy
  publication-title: Invest Ophthalmol Visual Sci
  doi: 10.1167/iovs.15-18736
– year: 2015
  ident: B36
  article-title: Semi-automated 2D bruch's membrane shape analysis in papilledema using spectral-domain optical coherence tomography
  publication-title: SPIE Med Imaging
– volume: 3
  start-page: 35
  year: 2018
  ident: B5
  article-title: The role of optical coherence tomography in neuro-ophthalmology
  publication-title: Ann Eye Sci
  doi: 10.21037/aes.2018.05.08
– volume: 52
  year: 2011
  ident: B14
  article-title: Macular ganglion cell–inner plexiform layer: Automated detection and thickness reproducibility with spectral domain–optical coherence tomography in glaucoma
  publication-title: Invest Ophthalmol Visual Sci
  doi: 10.1167/iovs.11-7962
– volume-title: Proceedings of the AAAI conference on artificial intelligence
  year: 2018
  ident: B18
  article-title: Multi-level variational autoencoder: Learning disentangled representations from grouped observations
  doi: 10.1609/aaai.v32i1.11867
– volume: 22
  year: 2016
  ident: B4
  article-title: Retinal ganglion cell layer thinning within one month of presentation for optic neuritis
  publication-title: Mult Scler
  doi: 10.1177/1352458515598020
– volume: 13
  start-page: 1
  year: 2024
  ident: B29
  article-title: Visualization of optic nerve structural patterns in papilledema using deep learning variational autoencoders
  publication-title: Trans Vision Sci Technol
  doi: 10.1167/tvst.13.1.13
– volume: 31
  year: 2011
  ident: B1
  article-title: Role of the macular optical coherence tomography scan in neuro-ophthalmology
  publication-title: J Neuro-Ophthalmology
  doi: 10.1097/WNO.0b013e318238b9cb
– volume: 12
  year: 2019
  ident: B15
  article-title: An introduction to variational autoencoders
  publication-title: Foundations Trends Mach Learn
  doi: 10.1561/2200000056
– volume: 1907
  year: 2019
  ident: B24
  article-title: Retina-VAE: Variationally decoding the spectrum of macular disease
  publication-title: ArXiv
– volume: 8
  year: 2022
  ident: B12
  article-title: Intraretinal layer segmentation using cascaded compressed u-nets
  publication-title: J Imaging
  doi: 10.3390/jimaging8050139
– volume: 58
  year: 2017
  ident: B37
  article-title: Peripapillary retinal pigment epithelium layer shape changes from acetazolamide treatment in the idiopathic intracranial hypertension treatment trial
  publication-title: Invest Ophthalmol Visual Sci
  doi: 10.1167/iovs.16-21089
– volume: 15
  year: 2024
  ident: B10
  article-title: Hybrid deep learning and optimal graph search method for optical coherence tomography layer segmentation in diseases affecting the optic nerve
  publication-title: Biomed Optics Express
  doi: 10.1364/BOE.516045
– volume-title: Proceedings of the 25th ACM SIGKDD international conference on knowledge discovery & Data mining
  year: 2019
  ident: B25
  article-title: Glaucoma progression prediction using retinal thickness via latent space linear regression
  doi: 10.1145/3292500.3330757
– start-page: 1
  volume-title: 2020 international joint conference on neural networks (IJCNN)
  year: 2020
  ident: B20
  article-title: Deep learning based unsupervised and semi-supervised classification for keratoconus
  doi: 10.1109/IJCNN48605.2020.9206694
– volume-title: Proceedings of the 22nd ACM SIGKDD international conference on knowledge discovery and data mining
  year: 2016
  ident: B30
  article-title: Xgboost: A scalable tree boosting system
  doi: 10.1145/2939672.2939785
– volume: 3
  year: 2020
  ident: B26
  article-title: Improving the structure-function relationship in glaucomatous visual fields by using a deep learning-based noise reduction approach
  publication-title: Ophthalmol Glaucoma
  doi: 10.1016/j.ogla.2020.01.001
– volume: 30
  start-page: 1
  year: 2022
  ident: B8
  article-title: Ms minute: Retinal optical coherence tomography for ms
  publication-title: Pratical Neurol
– volume-title: 2021 IEEE/CVF conference on computer vision and pattern recognition (CVPR).
  year: 2021
  ident: B35
  article-title: Soft-introvae: Analyzing and improving the introspective variational autoencoder
  doi: 10.1109/CVPR46437.2021.00437
– volume: 7
  year: 2013
  ident: B9
  article-title: Long-term retinal nerve fiber layer changes following nonarteritic anterior ischemic optic neuropathy
  publication-title: Clin Ophthalmol
  doi: 10.2147/OPTH.S42522
– ident: B32
  doi: 10.21105/joss.03021
– start-page: 1
  volume-title: Neural information processing systems (NeurIPS) workshop on learning disentangled representations
  year: 2018
  ident: B16
  article-title: Understanding disentangling in β-VAE
– volume: 13
  start-page: 20586
  year: 2023
  ident: B22
  article-title: Computer-aided diagnosis of keratoconus through VAE-augmented images using deep learning
  publication-title: Sci Rep
  doi: 10.1038/s41598-023-46903-5
– volume-title: Ophthalmic medical image analysis (OMIA)
  year: 2021
  ident: B6
  article-title: Representation and reconstruction of image-based structural patterns of glaucomatous defects using only two latent variables from a variational autoencoder
  doi: 10.1007/978-3-030-87000-3_17
– volume-title: 2017 IEEE Winter Conference on Applications of Computer Vision (WACV)
  year: 2016
  ident: B17
  article-title: Deep feature consistent variational autoencoder
– volume: 131
  start-page: 790
  year: 2024
  ident: B31
  article-title: Ophthalmic and systemic factors of acute nonarteritic anterior ischemic optic neuropathy in the quark207 treatment trial
  publication-title: Ophthalmology
  doi: 10.1016/j.ophtha.2024.01.011
– volume: 4
  start-page: 1
  year: 2024
  ident: B28
  article-title: Efficacy of smoothing algorithms to enhance detection of visual field progression in glaucoma
  publication-title: Ophthalmol Sci
  doi: 10.1016/j.xops.2023.100423
SSID ssj0002875949
Score 2.2887018
Snippet Glaucoma, optic neuritis (ON), and non-arteritic anterior ischemic optic neuropathy (NAION) produce distinct patterns of retinal ganglion cell (RGC) damage. We...
IntroductionGlaucoma, optic neuritis (ON), and non-arteritic anterior ischemic optic neuropathy (NAION) produce distinct patterns of retinal ganglion cell...
SourceID doaj
pubmedcentral
proquest
pubmed
crossref
SourceType Open Website
Open Access Repository
Aggregation Database
Index Database
Enrichment Source
StartPage 1497848
SubjectTerms glaucoma
non-arteritic anterior ischemic optic neuropathy (NAION)
Ophthalmology
optic neuritis (ON)
optical coherence tomography (OCT)
retinal ganglion cell (RGC) loss
variational autoencoder (VAE)
Title Quantifying the spatial patterns of retinal ganglion cell loss and progression in optic neuropathy by applying a deep learning variational autoencoder approach to optical coherence tomography
URI https://www.ncbi.nlm.nih.gov/pubmed/39963427
https://www.proquest.com/docview/3168026228
https://pubmed.ncbi.nlm.nih.gov/PMC11830743
https://doaj.org/article/3a90b2bedafc4242903109293caedec0
Volume 4
WOSCitedRecordID wos001422336000001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
hasFullText 1
inHoldings 1
isFullTextHit
isPrint
journalDatabaseRights – providerCode: PRVAON
  databaseName: DOAJ Directory of Open Access Journals
  customDbUrl:
  eissn: 2674-0826
  dateEnd: 99991231
  omitProxy: false
  ssIdentifier: ssj0002875949
  issn: 2674-0826
  databaseCode: DOA
  dateStart: 20210101
  isFulltext: true
  titleUrlDefault: https://www.doaj.org/
  providerName: Directory of Open Access Journals
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwrV1Lj9MwELZghRAXxJvyWA0SNxRtYruJcwTEigOsQAKpt8iPybZoN67adKX9dfw1Zpy0ahGCC9c4sa18Y8-MPfONEK8rX2kb0GWtVD7TpFEyF2i54zSXrsCWtuWE9Kfq7MzMZvWXvVJfHBM20AMPP-5E2Tp30mGwrdekT-rEZUlKylsM6JO3nlf1njP1Ix0ZVdNa10OWDHlh9Ukbl3OOnZSaNgdynbjgz54mSoT9f7Iyfw-W3NM-p_fE3dFshLfDdO-LG9g9ELc_jxfjD8XPrxvLYT-ctARk08GaI6Xpi2Xiz-zWEFvghEXu5dxy7m7sgE_t4YImB7YLkEK1BpoOWHQQaTPxkPguuWzxNbhr4OvuNISFgLiEsebEOVyRyz0eK4Ld9JHpMQOuYEtZDn0cOqR2H-dDjiE9vBwZsx-J76cfvr3_mI21GTJPDnWfSYW6tl7bytrCYDJ0DBlDJQGjc1RFWZJlqRLhnyzQGzTOllM7zVtXKI3qsTjqYodPBfjWt0WVO-eM0iFwGlirtTO-9CEYpyai2OLU-JG4nOtnXDTkwDC2TcK2YWybEduJeLP7ZjnQdvz17XcM_-5NptxOD0gQm1EQm38J4kS82gpPQ0uUEbQdxs264dpg5OpKSQM9GYRpNxT9tlJpWU2EORCzg7kctnSLeaIBJ9dQsQH47H_M_rm4I7myMcejqxfiqF9t8KW45a_6xXp1LG5WM3OcltgvOZwyyA
linkProvider Directory of Open Access Journals
openUrl ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=Quantifying+the+spatial+patterns+of+retinal+ganglion+cell+loss+and+progression+in+optic+neuropathy+by+applying+a+deep+learning+variational+autoencoder+approach+to+optical+coherence+tomography&rft.jtitle=Frontiers+in+ophthalmology&rft.au=Wang%2C+Jui-Kai&rft.au=Johnson%2C+Brett+A.&rft.au=Chen%2C+Zhi&rft.au=Zhang%2C+Honghai&rft.date=2025-02-03&rft.issn=2674-0826&rft.eissn=2674-0826&rft.volume=4&rft_id=info:doi/10.3389%2Ffopht.2024.1497848&rft.externalDBID=n%2Fa&rft.externalDocID=10_3389_fopht_2024_1497848
thumbnail_l http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=2674-0826&client=summon
thumbnail_m http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=2674-0826&client=summon
thumbnail_s http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=2674-0826&client=summon