Glaucoma detection in myopic eyes using deep learning autoencoder-based regions of interest
PurposeTo evaluate the diagnostic accuracy of a deep learning autoencoder-based model utilizing regions of interest (ROI) from optical coherence tomography (OCT) texture enface images for detecting glaucoma in myopic eyes.MethodsThis cross-sectional study included a total of 453 eyes from 315 partic...
Saved in:
| Main Authors: | , , , , , , , , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2025-08-01
|
| Series: | Frontiers in Ophthalmology |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fopht.2025.1624015/full |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1839582728047034368 |
|---|---|
| author | Christopher Bowd Akram Belghith Mark Christopher Makoto Araie Aiko Iwase Goji Tomita Kyoko Ohno-Matsui Hitomi Saito Hiroshi Murata Tsutomu Kikawa Kazuhisa Sugiyama Tomomi Higashide Atsuya Miki Atsuya Miki Toru Nakazawa Makoto Aihara Tae-Woo Kim Christopher Kai Shun Leung Robert N. Weinreb Linda M. Zangwill |
| author_facet | Christopher Bowd Akram Belghith Mark Christopher Makoto Araie Aiko Iwase Goji Tomita Kyoko Ohno-Matsui Hitomi Saito Hiroshi Murata Tsutomu Kikawa Kazuhisa Sugiyama Tomomi Higashide Atsuya Miki Atsuya Miki Toru Nakazawa Makoto Aihara Tae-Woo Kim Christopher Kai Shun Leung Robert N. Weinreb Linda M. Zangwill |
| author_sort | Christopher Bowd |
| collection | DOAJ |
| description | PurposeTo evaluate the diagnostic accuracy of a deep learning autoencoder-based model utilizing regions of interest (ROI) from optical coherence tomography (OCT) texture enface images for detecting glaucoma in myopic eyes.MethodsThis cross-sectional study included a total of 453 eyes from 315 participants from the multi-center "Swept-Source OCT (SS-OCT) Myopia and Glaucoma Study", composed of 268 eyes from 168 healthy individuals and 185 eyes from 147 glaucomatous individuals. All participants underwent swept-source optical coherence tomography (SS-OCT) imaging, from which texture enface images were constructed and analyzed. The study compared four methods: (1) global RNFL thickness, (2) texture enface image, (3) a single autoencoder model trained only on healthy eyes, and (4) a dual autoencoder model trained on both healthy and glaucomatous eyes. Diagnostic accuracy was assessed using the area under the receiver operating curves (AUROC) and precision recall curves (AUPRC).ResultsThe dual autoencoder model achieved the highest AUROC (95% CI) (0.92 [0.88, 0.95]), significantly outperforming the single autoencoder model trained only on healthy eyes (0.86 [0.83, 0.88], p = 0.01), the global RNFL thickness model (0.84 [0.80, 0.86], p = 0.003), and the texture enface model (0.83 [0.79, 0.85], p = 0.005). Using AUPRC (95% CI), the dual autoencoder model (0.86 [0.83, 0.89]) also outperformed the single autoencoder model trained only on healthy eyes (0.80 [0.78, 0.82], p = 0.02), the global RNFL thickness model (0.74 [0.70, 0.76], p = 0.001), and the texture enface model (0.71 [0.68, 0.73], p<0.001). No significant difference was observed between the global RNFL thickness measurement and the texture enface measurement (p = 0.47).DiscussionThe dual autoencoder model, which integrates reconstruction errors from both healthy and glaucomatous training data, demonstrated superior diagnostic accuracy compared to the single autoencoder model, global RNFL thickness and texture enface-based approaches. These findings suggest that deep learning models leveraging ROI-based reconstruction error from texture enface images may enhance glaucoma classification in myopic eyes, providing a robust alternative to conventional structural thickness metrics. |
| format | Article |
| id | doaj-art-ca50c03b8b8e4f6eba2057b51ed5fa8e |
| institution | Matheson Library |
| issn | 2674-0826 |
| language | English |
| publishDate | 2025-08-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Ophthalmology |
| spelling | doaj-art-ca50c03b8b8e4f6eba2057b51ed5fa8e2025-08-04T04:10:11ZengFrontiers Media S.A.Frontiers in Ophthalmology2674-08262025-08-01510.3389/fopht.2025.16240151624015Glaucoma detection in myopic eyes using deep learning autoencoder-based regions of interestChristopher Bowd0Akram Belghith1Mark Christopher2Makoto Araie3Aiko Iwase4Goji Tomita5Kyoko Ohno-Matsui6Hitomi Saito7Hiroshi Murata8Tsutomu Kikawa9Kazuhisa Sugiyama10Tomomi Higashide11Atsuya Miki12Atsuya Miki13Toru Nakazawa14Makoto Aihara15Tae-Woo Kim16Christopher Kai Shun Leung17Robert N. Weinreb18Linda M. Zangwill19Hamilton Glaucoma Center and Division of Ophthalmology Informatics and Data Science, Shiley Eye Institute, Viterbi Family Department of Ophthalmology, University of California (UC) San Diego, La Jolla, CA, United StatesHamilton Glaucoma Center and Division of Ophthalmology Informatics and Data Science, Shiley Eye Institute, Viterbi Family Department of Ophthalmology, University of California (UC) San Diego, La Jolla, CA, United StatesHamilton Glaucoma Center and Division of Ophthalmology Informatics and Data Science, Shiley Eye Institute, Viterbi Family Department of Ophthalmology, University of California (UC) San Diego, La Jolla, CA, United StatesKanto Central Hospital of the Mutual Aid Association of Public School Teachers, Tokyo, JapanTajimi Iwase Eye Clinic, Tajimi, JapanDepartment of Ophthalmology, Toho University Ohashi Medical Center, Tokyo, JapanDepartment of Ophthalmology and Visual Science, Tokyo Medical and Dental University, Tokyo, JapanDepartment of Ophthalmology, Graduate School of Medicine, The University of Tokyo, Tokyo, JapanCenter Hospital of the National Center for Global Health and Medicine, Tokyo, JapanR&D Division, Topcon Corporation, Tokyo, JapanDepartment of Ophthalmology, Kanazawa University Graduate School of Medical Sciences, Kanazawa, JapanDepartment of Ophthalmology, Kanazawa University Graduate School of Medical Sciences, Kanazawa, Japan0Department of Innovative Visual Science, Osaka University Graduate School of Medicine, Osaka, Japan1Department of Myopia Control Research, Aichi Medical University Medical School, Nagakute, Japan2Department of Ophthalmology, Tohoku University School of Medicine, Sendai, JapanDepartment of Ophthalmology, Graduate School of Medicine, The University of Tokyo, Tokyo, Japan3Department of Ophthalmology, Seoul National University College of Medicine, Seoul National University Bundang Hospital, Seongnam, Republic of Korea4Department of Ophthalmology, LKS Faculty of Medicine, The University of Hong Kong, Hong Kong, Hong Kong SAR, ChinaHamilton Glaucoma Center and Division of Ophthalmology Informatics and Data Science, Shiley Eye Institute, Viterbi Family Department of Ophthalmology, University of California (UC) San Diego, La Jolla, CA, United StatesHamilton Glaucoma Center and Division of Ophthalmology Informatics and Data Science, Shiley Eye Institute, Viterbi Family Department of Ophthalmology, University of California (UC) San Diego, La Jolla, CA, United StatesPurposeTo evaluate the diagnostic accuracy of a deep learning autoencoder-based model utilizing regions of interest (ROI) from optical coherence tomography (OCT) texture enface images for detecting glaucoma in myopic eyes.MethodsThis cross-sectional study included a total of 453 eyes from 315 participants from the multi-center "Swept-Source OCT (SS-OCT) Myopia and Glaucoma Study", composed of 268 eyes from 168 healthy individuals and 185 eyes from 147 glaucomatous individuals. All participants underwent swept-source optical coherence tomography (SS-OCT) imaging, from which texture enface images were constructed and analyzed. The study compared four methods: (1) global RNFL thickness, (2) texture enface image, (3) a single autoencoder model trained only on healthy eyes, and (4) a dual autoencoder model trained on both healthy and glaucomatous eyes. Diagnostic accuracy was assessed using the area under the receiver operating curves (AUROC) and precision recall curves (AUPRC).ResultsThe dual autoencoder model achieved the highest AUROC (95% CI) (0.92 [0.88, 0.95]), significantly outperforming the single autoencoder model trained only on healthy eyes (0.86 [0.83, 0.88], p = 0.01), the global RNFL thickness model (0.84 [0.80, 0.86], p = 0.003), and the texture enface model (0.83 [0.79, 0.85], p = 0.005). Using AUPRC (95% CI), the dual autoencoder model (0.86 [0.83, 0.89]) also outperformed the single autoencoder model trained only on healthy eyes (0.80 [0.78, 0.82], p = 0.02), the global RNFL thickness model (0.74 [0.70, 0.76], p = 0.001), and the texture enface model (0.71 [0.68, 0.73], p<0.001). No significant difference was observed between the global RNFL thickness measurement and the texture enface measurement (p = 0.47).DiscussionThe dual autoencoder model, which integrates reconstruction errors from both healthy and glaucomatous training data, demonstrated superior diagnostic accuracy compared to the single autoencoder model, global RNFL thickness and texture enface-based approaches. These findings suggest that deep learning models leveraging ROI-based reconstruction error from texture enface images may enhance glaucoma classification in myopic eyes, providing a robust alternative to conventional structural thickness metrics.https://www.frontiersin.org/articles/10.3389/fopht.2025.1624015/fullglaucomamyopiaoptical coherence tomographydeep learningartificial intelligencediagnosis |
| spellingShingle | Christopher Bowd Akram Belghith Mark Christopher Makoto Araie Aiko Iwase Goji Tomita Kyoko Ohno-Matsui Hitomi Saito Hiroshi Murata Tsutomu Kikawa Kazuhisa Sugiyama Tomomi Higashide Atsuya Miki Atsuya Miki Toru Nakazawa Makoto Aihara Tae-Woo Kim Christopher Kai Shun Leung Robert N. Weinreb Linda M. Zangwill Glaucoma detection in myopic eyes using deep learning autoencoder-based regions of interest Frontiers in Ophthalmology glaucoma myopia optical coherence tomography deep learning artificial intelligence diagnosis |
| title | Glaucoma detection in myopic eyes using deep learning autoencoder-based regions of interest |
| title_full | Glaucoma detection in myopic eyes using deep learning autoencoder-based regions of interest |
| title_fullStr | Glaucoma detection in myopic eyes using deep learning autoencoder-based regions of interest |
| title_full_unstemmed | Glaucoma detection in myopic eyes using deep learning autoencoder-based regions of interest |
| title_short | Glaucoma detection in myopic eyes using deep learning autoencoder-based regions of interest |
| title_sort | glaucoma detection in myopic eyes using deep learning autoencoder based regions of interest |
| topic | glaucoma myopia optical coherence tomography deep learning artificial intelligence diagnosis |
| url | https://www.frontiersin.org/articles/10.3389/fopht.2025.1624015/full |
| work_keys_str_mv | AT christopherbowd glaucomadetectioninmyopiceyesusingdeeplearningautoencoderbasedregionsofinterest AT akrambelghith glaucomadetectioninmyopiceyesusingdeeplearningautoencoderbasedregionsofinterest AT markchristopher glaucomadetectioninmyopiceyesusingdeeplearningautoencoderbasedregionsofinterest AT makotoaraie glaucomadetectioninmyopiceyesusingdeeplearningautoencoderbasedregionsofinterest AT aikoiwase glaucomadetectioninmyopiceyesusingdeeplearningautoencoderbasedregionsofinterest AT gojitomita glaucomadetectioninmyopiceyesusingdeeplearningautoencoderbasedregionsofinterest AT kyokoohnomatsui glaucomadetectioninmyopiceyesusingdeeplearningautoencoderbasedregionsofinterest AT hitomisaito glaucomadetectioninmyopiceyesusingdeeplearningautoencoderbasedregionsofinterest AT hiroshimurata glaucomadetectioninmyopiceyesusingdeeplearningautoencoderbasedregionsofinterest AT tsutomukikawa glaucomadetectioninmyopiceyesusingdeeplearningautoencoderbasedregionsofinterest AT kazuhisasugiyama glaucomadetectioninmyopiceyesusingdeeplearningautoencoderbasedregionsofinterest AT tomomihigashide glaucomadetectioninmyopiceyesusingdeeplearningautoencoderbasedregionsofinterest AT atsuyamiki glaucomadetectioninmyopiceyesusingdeeplearningautoencoderbasedregionsofinterest AT atsuyamiki glaucomadetectioninmyopiceyesusingdeeplearningautoencoderbasedregionsofinterest AT torunakazawa glaucomadetectioninmyopiceyesusingdeeplearningautoencoderbasedregionsofinterest AT makotoaihara glaucomadetectioninmyopiceyesusingdeeplearningautoencoderbasedregionsofinterest AT taewookim glaucomadetectioninmyopiceyesusingdeeplearningautoencoderbasedregionsofinterest AT christopherkaishunleung glaucomadetectioninmyopiceyesusingdeeplearningautoencoderbasedregionsofinterest AT robertnweinreb glaucomadetectioninmyopiceyesusingdeeplearningautoencoderbasedregionsofinterest AT lindamzangwill glaucomadetectioninmyopiceyesusingdeeplearningautoencoderbasedregionsofinterest |