Automated diagnostics of epiretinal membrane on OCT images using deep learning algorithms

Purpose: to develop and evaluate the effectiveness of a neural network model for automatic segmentation of epiretinal membrane (ERM) in optical coherence tomography (OCT) images.

Materials and methods. The study includes 322 labeled macular OCT scans with signs of ERM: 167 from the private dataset of the “Professorskaya Plus” clinic and 155 from the public OCTDL dataset. Five architectures were selected for comparison: U-Net, Attention U-Net, TransUNet, LOCTSeg, and Tiny-UNet. Initial annotations were generated using a baseline U-Net model and underwent expert clinical validation. Segmentation performance was assessed using Dice coefficient and Intersection over Union (IoU). Annotation quality was ensured by three experienced ophthalmologists with over 10 years of clinical practice.

Results. All models demonstrated comparable Dice and IoU scores, with no statistically significant differences. Tiny-UNet showed the best balance of accuracy and computational efficiency: 570K parameters, 5× faster training per epoch than U-Net, and total training time of only 20 min. Segmentation accuracy reached Dice = 86.1 %, IoU = 78.6 %.

Conclusion. Tiny-UNet appears to be the optimal architecture for ERM segmentation tasks, offering high accuracy with minimal computational requirements. Its efficiency makes it suitable for clinical deployment, including in mobile and cloud-based telemedicine platforms.

1. Dupas B, Tadayoni R, Gaudric A. Epiretinal membranes. J Fr Ophthalmol. Российский офтальмологический журнал. 2025; 38, 9: 861–75. doi: 10.1016/j.jfo.2015.08.004

2. Fung AT, Galvin J, Tran T. Epiretinal membrane: A review. Clin Exp Ophthalmol. 2021; 49, 3: 289–308. doi: 10.1111/ceo.13914

3. Vallejo-Garcia JL, Romano M, Pagano L, et al. OCT changes of idiopathic epiretinal membrane after cataract surgery. Int J Retina Vitreous. 2020; 6, 37. doi: 10.1186/s40942-020-00239-8

4. Mitchell P, Smith W, Chey T, et al. Prevalence and associations of epiretinal membranes. The Blue Mountains Eye Study. Ophthalmology. 1997 Jun; 104, 6: 1033–40. doi: 10.1016/s0161–6420(97)30190–0

5. Gale MJ, Scruggs BA, Flaxel CJ, Diabetic eye disease: A review of screening and management recommendations. Clin Exp Ophthalmol. 2021 Mar; 49 (2): 128–45. doi: 10.1111/ceo.13894

6. Govetto A, Lalane RA, Sarraf D, et al. Insights into epiretinal membranes: Presence of ectopic inner foveal layers and a new optical coherence tomography staging scheme. Am J Ophthalmol. 2017; 175: 99–113. doi: 10.1016/j.ajo.2016.12.006

7. Shpak A.A., Zhuravlev A.S., Kolesnik S.V., et al. Prognostic criteria for functional outcomes of surgical treatment of idiopathic epiretinal fibrosis based on optical coherence tomography. Fyodorov journal of ophthalmic surgery. 2024; 2 (139): 109–15 (In Russ.)]. https://doi.org/10.25276/0235-4160-2024-2-109-115

8. Stevenson W, Ponce CMP, Agarwal DR, et al. Epiretinal membrane: optical coherence tomography-based diagnosis and classification. Clin Ophthalmol. 2016; 10: 527–34. doi: 10.2147/OPTH.S97722

9. Baamonde S, de Moura J, Novo J, et al. Automatic identification and intuitive map representation of the epiretinal membrane presence in 3D OCT volumes. Sensors. 2019; 19 (23). doi: 10.3390/s19235269

10. Gende M, de Moura J, Novo J, et al. Fully automatic epiretinal membrane segmentation in OCT scans using convolutional networks. AI Applications for Disease Diagnosis and Treatment, IGI Global Scientific Publishing. 2022; 88–121. doi: 10.4018/978-1-6684-2304-2.ch004

11. Tang Y, Gao X, Wang W, et al. Automated detection of epiretinal membranes in OCT images using deep learning. Ophthalmic Res. 2023; 66 (1): 238–46. doi: 10.1159/000525929

12. Parra-Mora E, da Silva Cruz LA. LOCTseg: A lightweight fully convolutional network for end-to-end optical coherence tomography segmentation. Comput Biol Med. 2022 Nov; 150: 106174. doi: 10.1016/j.compbiomed.2022.106174

13. Katalevskaya E.A., Sizov A.Yu., Gilemzyanova L.I. Artificial intelligence algorithm for segmentation of pathological structures on retinal optical coherence tomography scans. Rossijskij zhurnal telemediciny i elektronnogo zdravoohraneniya. 2022; 8 (3): 21–7 (In Russ.)]. https://doi.org/10.29188/2712-9217-2022-8-3-21-27

14. Chen J, Lu Y, Yu Q, et al. TransUNet: Transformers make strong encoders for medical image segmentation. 2021 Feb; arXiv: arXiv:2102.04306. doi: 10.48550/arXiv.2102.04306

15. Kulyabin M, Zhdanov A, Nikiforova A, et al. OCTDL: Optical coherence tomography dataset for image-based deep learning methods. Sci Data. 2024 Apr; 11 (1): 365. doi: 10.1038/s41597-024-03182-7

16. Kulyabin M, Zhdanov A, Pershin A, et al. Segment anything in optical coherence tomography: SAM 2 for volumetric segmentation of retinal biomarkers. Bioengineering. 2024 Sep; 11 (9): 940. doi: 10.3390/bioengineering11090940