Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2025 Aug 4:5:1624015.
doi: 10.3389/fopht.2025.1624015. eCollection 2025.

Glaucoma detection in myopic eyes using deep learning autoencoder-based regions of interest

Christopher Bowd  1 , Akram Belghith  1 , Mark Christopher  1 , Makoto Araie  2 , Aiko Iwase  3 , Goji Tomita  4 , Kyoko Ohno-Matsui  5 , Hitomi Saito  6 , Hiroshi Murata  7 , Tsutomu Kikawa  8 , Kazuhisa Sugiyama  9 , Tomomi Higashide  9 , Atsuya Miki  10   11 , Toru Nakazawa  12 , Makoto Aihara  6 , Tae-Woo Kim  13 , Christopher Kai Shun Leung  14 , Robert N Weinreb  1 , Linda M Zangwill  1
Affiliations

Glaucoma detection in myopic eyes using deep learning autoencoder-based regions of interest

Christopher Bowd et al. Front Ophthalmol (Lausanne). .

Abstract

Purpose: To 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.

Methods: This 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).

Results: The 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).

Discussion: The 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.

Keywords: artificial intelligence; classification; deep learning; diagnosis; glaucoma; myopia; optical coherence tomography.

PubMed Disclaimer

Conflict of interest statement

Author TK was employed by the company Topcon Corporation. HS, AM, TN, TK, CL, RW, and LZ all receive research support from Topcon. The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Figures

Figure 1
Figure 1
Example of glaucomatous eyes, showing (a) RNFL (retinal nerve fiber layer) thickness map, (b) texture image, (c) SLO (scanning laser ophthalmoscopy) image, (d) dual autoencoder ROI-based map, and (e) single autoencoder ROI-based map.
See this image and copyright information in PMC

Similar articles

See all similar articles

References

    1. Wang YX, Yang H, Wei CC, Xu L, Wei WB, Jonas JB. High myopia as risk factor for the 10-year incidence of open-angle glaucoma in the Beijing Eye Study. Br J Ophthalmol. (2023) 107:935–40. doi: 10.1136/bjophthalmol-2021-320644, PMID: - DOI - PMC - PubMed
    1. Leske MC, Connell AM, Schachat AP, Hyman L, The Barbados Eye Study . Prevalence of open angle glaucoma. Arch Ophthalmol. (1994) 112:821–9. - PubMed
    1. Xu L, Wang Y, Wang S, Wang Y, Jonas JB. High myopia and glaucoma susceptibility the Beijing Eye Study. Ophthalmology. (2007) 114:216–20. doi: 10.1016/j.ophtha.2006.06.050, PMID: - DOI - PubMed
    1. Mitchell P, Hourihan F, Sandbach J, Wang JJ. The relationship between glaucoma and myopia: the Blue Mountain Eye Study. Ophthalmology. (1999) 106:2010–5. - PubMed
    1. Holden BA, Fricke TR, Wilson DA, Jong M, Naidoo KS, Sankaridurg P, et al. Global prevalence of myopia and high myopia and temporal trends from 2000 through 2050. Ophthalmology. May. (2016) 123:1036–42. doi: 10.1016/j.ophtha.2016.01.006, PMID: - DOI - PubMed

LinkOut - more resources