A combined software method for analyzing filtering bleb vascularization from tomographic images using computer vision techniques

Petrov S. Yu.¹

Solodovnikov V. I.²

Milash S. V.¹

Markelova O. I.¹

Ephieva A. D.¹

Pavlov K. V.¹

Karakotova N. V.³

1Helmholtz National Medical Research Center of Eye Diseases, Moscow, Russia.
2Design Information Technologies Center Russian Academy of Sciences, Moscow Region, Odintsovo, Russia.
3Bauman Moscow State Technical University, Moscow, Russia.

Markelova O. I.; e-mail: levinaoi@mail.ru

None declared.

The study had no sponsorship.

The state of the filtration bleb is a significant prognostic factor for the outcome of glaucoma surgery; however, existing methods for its assessment are subjective. The aim of this study was to develop an automated method for the objective analysis of filtering bleb images. A dataset of 100 patients with primary open-angle glaucoma who underwent trabeculectomy was used. This dataset included biomicroscopic photographs and OCT-angiographic tomograms performed preoperatively and at 1, 2, and 6 weeks postoperatively. Vascular density and tortuosity (from tomographic images), as well as hyperemia (from color photographs), were calculated programmatically using the corresponding formulas. A convolutional neural network with the U-Net architecture was trained using the calculated parameters in the resulting «OCT-A image – FP photo» pairs to automate the subsequent hyperemia calculation based on the tomographic image using an artificial intelligence tool. This resulted in the creation of specialized software that enables the quantitative assessment of key parameters of the filtration bleb state. The developed method showed high accuracy of the U-Net model during training (88.24%) and subsequent retraining (85.31%). Thus, the proposed solution provides a tool for objective monitoring of the postoperative period, which can contribute to the optimization of patient management strategies and improved prognosis for glaucoma surgery.

filtering bleb, glaucoma, computer vision, imaging segmentation, OCT-angiography, convolutional neural network, trabeculectomy

10.29234/2308-9113-2025-13-4-1-16

1. Quigley H.A., Broman A.T. The number of people with glaucoma worldwide in 2010 and 2020. Br J Ophthalmol 2006; 90(3): 262-267. doi:10.1136/bjo.2005.081224

2. Sotimehin A.E., Ramulu P.Y. Measuring Disability in Glaucoma. J Glaucoma 2018; 27(11): 939-949. doi:10.1097/IJG.0000000000001068

3. Tielsch J.M., Katz J., Singh K., Quigley H.A., Gottsch J.D., Javitt J., Sommer A. A population-based evaluation of glaucoma screening: the Baltimore Eye Survey. Am J Epidemiol 1991; 134(10): 1102-1110. doi:10.1093/oxfordjournals.aje.a116013

4. Tham Y.C., Li X., Wong T.Y., Quigley H.A., Aung T., Cheng C.Y. Global prevalence of glaucoma and projections of glaucoma burden through 2040: a systematic review and meta-analysis. Ophthalmology 2014; 121(11): 2081-2090. doi:10.1016/j.ophtha.2014.05.013

5. Клинические рекомендации «Глаукома первичная открытоугольная» (утверждены МЗ РФ, 17.08.2024).

6. Picht G., Grehn F. Classification of filtering blebs in trabeculectomy: biomicroscopy and functionality. Curr Opin Ophthalmol 1998; 9(2): 2-8. doi:10.1097/00055735-199804000-00002

7. Broadway D.C., Bloom P.A., Bunce C., Thiagarajan M., Khaw P.T. Needle revision of failing and failed trabeculectomy blebs with adjunctive 5-fluorouracil: survival analysis. Ophthalmology 2004; 111(4): 665-673. doi:10.1016/j.ophtha.2003.07.009

8. Feyi-Waboso A., Ejere H.O. Needling for encapsulated trabeculectomy filtering blebs. Cochrane Database Syst Rev 2012; 2012(8): CD003658. doi:10.1002/14651858.CD003658.pub3

9. Cantor L.B., Mantravadi A., WuDunn D., Swamynathan K., Cortes A. Morphologic classification of filtering blebs after glaucoma filtration surgery: the Indiana Bleb Appearance Grading Scale. J Glaucoma 2003; 12(3): 266-271. doi:10.1097/00061198-200306000-00015

10. Klink T., Kann G., Ellinger P., Klink J., Grehn F., Guthoff R. The prognostic value of the wuerzburg bleb classification score for the outcome of trabeculectomy. Ophthalmologica 2011; 225(1): 55-60. doi:10.1159/000314717

11. Smith M., Chipman M.L., Trope G.E., Buys Y.M. Correlation between the indiana bleb appearance grading scale and intraocular pressure after phacotrabeculectomy. J Glaucoma 2009; 18(3): 217-219. doi:10.1097/IJG.0b013e31817d23e0

12. Wells A.P., Ashraff N.N., Hall R.C., Purdie G. Comparison of two clinical Bleb grading systems. Ophthalmology 2006; 113(1): 77-83. doi:10.1016/j.ophtha.2005.06.037

13. Еричев В.П., Новиков И.А., Хачатрян Г.К., и др. Объем фильтрационной подушки в прогнозировании эффективности антиглаукомной операции: предварительные результаты. Национальный журнал глаукома. 2024; 23(4):3-10. doi: 10.53432/2078-4104-2024-23-4-3-10.

14. Beer T.W., Baldwin H.C., Goddard J.R., Gallagher P.J., Wright D.H. Angiogenesis in pathological and surgical scars. Hum Pathol 1998; 29(11): 1273-1278. doi:10.1016/s0046-8177(98)90256-8

15. Shaunak S., Thomas S., Gianasi E., Godwin A., Jones E., Teo I., Mireskandari K., Luthert P., Duncan R., Patterson S., Khaw P., Brocchini S. Polyvalent dendrimer glucosamine conjugates prevent scar tissue formation. Nat Biotechnol 2004; 22(8): 977-984. doi:10.1038/nbt995

16. Brown N.J., Smyth E.A., Cross S.S., Reed M.W. Angiogenesis induction and regression in human surgical wounds. Wound Repair Regen 2002; 10(4): 245-251. doi:10.1046/j.1524-475x.2002.10408.x

17. Шмырева В.Ф., Петров С.Ю., Антонов А.А., Сипливый В.И., Стратонников А.А., Савельева Т.А., Шевчик С.А., Рябова А.В. Метод оценки оксигенации субконъюнктивального сосудистого русла с помощью спектроскопии отраженного света (экспериментальное исследование). Глаукома 2008; 2: 9-14.

18. Шмырева В.Ф., Петров С.Ю., Антонов А.А., Стратонников А.А., Савельева Т.А., Шевчик С.А., Рябова А.В., Урываев Ю.В. Исследование метаболизма тканей переднего отрезка глаза по уровню оксигенации гемоглобина в венозном русле при первичной открытоугольной глаукоме. Глаукома 2008; 3: 3-10.

19. Hayek S., Labbe A., Brasnu E., Hamard P., Baudouin C. Optical Coherence Tomography Angiography Evaluation of Conjunctival Vessels During Filtering Surgery. Transl Vis Sci Technol 2019; 8(4): 4. doi:10.1167/tvst.8.4.4

20. Kido A., Akagi T., Ikeda H.O., Kameda T., Suda K., Miyake M., Hasegawa T., Numa S., Tsujikawa A. Longitudinal changes in complete avascular area assessed using anterior segmental optical coherence tomography angiography in filtering trabeculectomy bleb. Sci Rep 2021; 11(1): 23418. doi:10.1038/s41598-021-02871-2

21. Liu Z., Wang H., Jiang H., Gameiro G.R., Wang J. Quantitative analysis of conjunctival microvasculature imaged using optical coherence tomography angiography. Eye Vis (Lond) 2019; 6(5. doi:10.1186/s40662-019-0130-9

22. Mastropasqua R., Brescia L., Di Antonio L., Guarini D., Giattini D., Zuppardi E., Agnifili L. Angiographic biomarkers of filtering bleb function after XEN gel implantation for glaucoma: an optical coherence tomography-angiography study. Acta Ophthalmol 2020; 98(6): e761-e767. doi:10.1111/aos.14371

23. Mastropasqua R., Fasanella V., Agnifili L., Curcio C., Ciancaglini M., Mastropasqua L. Anterior segment optical coherence tomography imaging of conjunctival filtering blebs after glaucoma surgery. Biomed Res Int 2014; 2014(610623. doi:10.1155/2014/610623

24. Seo J.H., Kim Y.A., Park K.H., Lee Y. Evaluation of Functional Filtering Bleb Using Optical Coherence Tomography Angiography. Transl Vis Sci Technol 2019; 8(3): 14. doi:10.1167/tvst.8.3.14

25. Seo J.H., Lee Y., Shin J.H., Kim Y.A., Park K.H. Comparison of conjunctival vascularity changes using optical coherence tomography angiography after trabeculectomy and phacotrabeculectomy. Graefes Arch Clin Exp Ophthalmol 2019; 257(10): 2239-2255. doi:10.1007/s00417-019-04412-0

26. Yin X., Cai Q., Song R., He X., Lu P. Relationship between filtering bleb vascularization and surgical outcomes after trabeculectomy: an optical coherence tomography angiography study. Graefes Arch Clin Exp Ophthalmol 2018; 256(12): 2399-2405. doi:10.1007/s00417-018-4136-0

27. de Carlo T.E., Romano A., Waheed N.K., Duker J.S. A review of optical coherence tomography angiography (OCTA). Int J Retina Vitreous 2015; 1(5. doi:10.1186/s40942-015-0005-8

28. Lopez M.R., Tyrsa V., Sergiyenko O. Machine Vision: Approaches and Limitations. Computer Vision2008. p. 395-411.

29. Parihar V.R., Boveiri H.R. A Survey and Comparative Analysis on Image Segmentation Techniques. Image Segmentation: A Guide to Image Mining. 2018. p. 1-15.

30. Ghosh A., Sultana F., Sufian A., Chakrabarti A. Fundamental Concepts of Convolutional Neural Network. Recent Trends and Advances in Artificial Intelligence and Internet of Things. 2020. p. 519-567.

31. Ronneberger O., Fischer.P., T. B. U-Net: Convolutional Networks for Biomedical Image Segmentation. International Conference on Medical Image Computing and Computer-Assisted Intervention. 2016. p. 234-241.

32. Wei P., Beer M. Regression Models for Machine Learning. Machine Learning in Modeling and Simulation. 2023.