Glaucoma Detection in Myopic Eyes Using Deep Learning Autoencoder-Based Regions of Interest

Christopher Bowd¹Akram Belghith¹Mark Chrstopher¹Makoto Araie²Aiko Iwase³Gouji Tomita⁴Kyoko Ohno-Matsui⁵Hitomi Saito⁶Hiroshi Murata⁷Tsutomu Kikawa⁸Kazuhisa Sugiyama⁹Tomomi Higashide⁹Atsuya Miki^10,11Toru Nakazawa¹²Makoto Aihara⁶Tae-Woo Kim¹³Christopher Kai-Shun Leung¹⁴Robert N Weinreb¹Linda M Zangwill^1*

• ¹Viterbi Family Department of Ophthalmology, University of California, San Diego, La Jolla, California, United States
• ²Kanto Central Hospital of the Mutual Aid Association of Public School Teachers, Setagaya, Tōkyō, Japan
• ³Tajimi Iwase Eye Clinic, Tajima, Japan
• ⁴Department of Ophthalmology, Toho University Ohashi Medical Center, Tokyo, Japan
• ⁵Department of Ophthalmology, Tokyo Medical and Dental University, Tokyo, Japan
• ⁶Department of Ophthalmology, Graduate School of Medicine, The University of Tokyo, Tokyo, Japan
• ⁷Central Hospital of the National Center for Global Health and Medicine, Tokyo, Japan
• ⁸R&D Division, Topcon Corporation, Tokyo, Japan
• ⁹Department of Ophthalmology, Graduate School of Medical Sciences, Kanazawa University, Kanazawa, Ishikawa, Japan
• ¹⁰Department of Innovative Visual Science, Osaka City University, Osaka, Japan
• ¹¹Department of Myopia Control Research, Aichi Medical University, Nagakute, Aichi, Japan
• ¹²Department of Ophthalmology, Graduate School of Education, Tohoku University, Sendai, Japan
• ¹³Department of Ophthalmology, Seoul National University, Seoul, Republic of Korea
• ¹⁴Department of Ophthalmology, LKS Faculty of Medicine, University of Hong Kong, Hong Kong, China

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 multicenter "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 sweptsource 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.