Journal of Ophthalmic and Vision Research

Published date: Jun 18 2025

Journal Title: Journal of Ophthalmic and Vision Research

Issue title: ‎Volume 20 - 2025

Pages: 1 - 17

DOI: 10.18502/jovr.v20.17716

Hazem Abdelmotaalhazem@aun.edu.egDepartment of Ophthalmology, Assiut University, Assiut

Rossen Mihaylov Hazarbasanovhazarbassanov@gmail.comHospital de Olhos-CRO, Guarulhos, SP

Ramin Saloutisalouti.erc@gmail.comPoostchi Ophthalmology Research Center, Shiraz University of Medical Sciences, Shiraz

M. Hossein Nowroozzadehnorozzadeh@gmail.comPoostchi Ophthalmology Research Center, Shiraz University of Medical Sciences, Shiraz

Suphi Taneritaneri@refraktives-zentrum.deZentrum für Refraktive Chirurgie, Muenster

Ali H. Al-Timemyali.altimemy@kecbu.uobaghdad.edu.iqBiomedical Engineering Department, Al-Khwarizmi College of Engineering, University of Baghdad

Alexandru Lavriclavricalexandru@gmail.comComputers, Electronics and Automation Department, Stefan cel Mare University of Suceava

Hidenori Takahashitakahashi.hidenori8@gmail.comDepartment of Ophthalmology, Jichi Medical University, Tochigi

Siamak Yousefisiamak.yousefi@uthsc.eduDepartment of Ophthalmology, University of Tennessee Health Science Center, Memphis, TN

Purpose: To assess the performance of a hybrid Transformer-based convolutional neural network (CNN) model for automated detection of keratoconus in stand-alone Scheimpflug-based dynamic corneal deformation videos (DCDVs).
Methods: We used transfer learning for feature extraction from DCDVs. These feature maps were augmented by self-attention to model long-range dependencies before classification to identify keratoconus directly. Model performance was evaluated by objective accuracy metrics based on DCDVs from two independent cohorts with 275 and 546 subjects.
Results: The model’s sensitivity and specificity in detecting keratoconus were 93% and 84%, respectively. The AUC of the keratoconus probability score based on the external validation database was 0.97.
Conclusion: The hybrid Transformer-based model was highly sensitive and specific in discriminating normal from keratoconic eyes using DCDV(s) at levels that may prove useful in clinical practice.

Keywords: Artificial Intelligence, Corneal Imaging, Deep Learning, Keratoconus, Scheimpflug Imaging

1. Rabinowitz YS. Keratoconus. Surv Ophthalmol 1998;42:297–319.

2. Labiris G, Giarmoukakis A, Sideroudi H, Gkika M, Fanariotis M, Kozobolis V. Impact of keratoconus, cross-linking and cross-linking combined with photorefractive keratectomy on self-reported quality of life. Cornea 2012;31:734–739.

3. Fernández Pérez J, Valero Marcos A, Martínez Peña FJ. Early diagnosis of keratoconus: What difference is it making? Br J Ophthalmol 2014;98:1465–1466.

4. Piñero DP. Technologies for anatomical and geometric characterization of the corneal structure and anterior segment: A review. Semin Ophthalmol 2015;30:161–170.

5. Scarcelli G, Besner S, Pineda R, Yun SH. Biomechanical characterization of keratoconus corneas ex vivo with Brillouin microscopy. Invest Ophthalmol Vis Sci 2014;55:4490–4495.

6. Roberts CJ, Dupps WJ Jr. Biomechanics of corneal ectasia and biomechanical treatments. J Cataract Refract Surg 2014;40:991–998.

7. Roberts CJ, Mahmoud AM, Bons JP, Hossain A, Elsheikh A, Vinciguerra R, et al. Introduction of two novel stiffness parameters and interpretation of air puff-induced biomechanical deformation parameters with a dynamic Scheimpflug analyzer. J Refract Surg 2017;33:266–273.

8. Vinciguerra R, Ambrósio R Jr, Elsheikh A, Roberts CJ, Lopes B, Morenghi E, et al. Detection of keratoconus with a new biomechanical index. J Refract Surg 2016;32:803–810.

9. Ambrósio R Jr, Lopes BT, Faria-Correia F, Salomão MQ, Bühren J, Roberts CJ, et al. Integration of Scheimpflugbased corneal tomography and biomechanical assessments for enhancing ectasia detection. J Refract Surg 2017;33:434–443.

10. Ali NQ, Patel DV, McGhee CN. Biomechanical responses of healthy and keratoconic corneas measured using a noncontact Scheimpflug-based tonometer. Invest Ophthalmol Vis Sci 2014;55:3651–3659.

11. Bichen W, Chenfeng X, Dai X, Wan A, Zhang P, Tomizuka M, et al. Visual Transformers: Token-based image representation and processing for computer vision. ArXiv; 2020. https://doi.org/10.48550/arXiv.2006.03677

12. Vaswani A, Shazeer N, Parmar N, Uszkoreit J, Jones L, Gomez AN, et al. Attention is all you need. ArXiv; 2017. https://doi.org/10.48550./arXiv.1706.0376213. Bello I, Zoph B, Vaswani A, Shlens J, Le QV. Attention augmented convolutional networks. ArXiv; 2019. https://doi.org/10.48550./arXiv.1904.09925

14. Dosovitskiy A, Beyer L, Kolesnikov A, Weissenborn D, Zhai X, Unterthiner T, et al. An image is worth 16x16 words: Transformers for image recognition at scale. ArXiv; 2020. https://doi.org/10.48550./arXiv.2010.11929

15. Arnab A, Dehghani M, Heigold G, Sun C, Lučić M, Schmid C. ViViT: A video vision Transformer. ArXiv; 2021. https://doi.org/10.48550./arXiv.2103.15691

16. Krumeich JH, Daniel J, Knülle A. Live-epikeratophakia for keratoconus. J Cataract Refract Surg 1998;24:456–463.

17. Joda AA, Shervin MM, Kook D, Elsheikh A. Development and validation of a correction equation for Corvis tonometry. Comput Methods Biomech Biomed Engin 2016;19:943–953.

18. Xu X, Lin J, Tao Y, Wang X. An improved DenseNet method based on transfer learning for fundus medical images. 7th International Conference on Digital Home (ICDH) 2018:137–140.

19. Zhang C, Benz P, Argaw DM, Lee S, Kim J, Rameau F, et al. ResNet or DenseNet? Introducing dense shortcuts to ResNet. ArXiv; 2020. https://doi.org/10.48550./arXiv.2010.12496

20. Deng J, Dong W, Socher R, Li LJ, Li K, Fei-Fei L. Imagenet: A large-scale hierarchical image database. IEEE Conference on Computer Vision and Pattern Recognition 2009:248–255.

21. Goodfellow I, Bengio Y, Courville A. Deep learning. MIT Press; 2016.

22. Ruder S. An overview of gradient descent optimization algorithms. ArXiv; 2016. https://doi.org/10.48550./arXiv.1609.04747.

23. Powers DM. Evaluation: From precision, recall, and Fmeasure to ROC, informedness, markedness & correlation. J Mach Learn Technol 2011;2:37–63.

24. Webb GI. Naïve Bayes. In: Sammut C, Webb GI, editors. Encyclopedia of machine learning. Springer; 2011.

25. Sun X, Xu W. Fast implementation of DeLong’s algorithm for comparing the areas under correlated receiver operating characteristic curves. IEEE Signal Process Lett 2014;21:1389–1393.

26. Jolliffe IT, Cadima J. Principal component analysis: A review and recent developments. Philos Trans A Math Phys Eng Sci 2016;374:20150202.

27. Zhang R, Isola P, Efros AA, Shechtman E, Wang O. The unreasonable effectiveness of deep features as a perceptual metric. ArXiv; 2018. https://doi.org/10.1109/CVPR.2018.00068

28. Oliphant TE. Python for scientific computing. Comput Sci Eng 2007;9:10–20.

29. Pedregosa F, Varoquaux G, Gramfort A, Gramfort A, Michel V, Thirion B, et al. Scikit-learn: Machine learning in Python. J Mach Learn Res 2011;12:2825–2830.

30. Rahimzadeh M, Attar A. A modified deep convolutional neural network for detecting COVID-19 and pneumonia from chest X-ray images based on the concatenation of Xception and ResNet50V2. Inform Med Unlocked 2020;19:100360.

31. Smadja D, Touboul D, Cohen A, Doveh E, Santhiago MR, Mello GR, et al. Detection of subclinical keratoconus using an automated decision tree classification. Am J Ophthalmol 2013;156:237–246.e1.

32. Shi C, Wang M, Zhu T, Zhang Y, Ye Y, Jiang J, et al. Machine learning helps improve diagnostic ability of subclinical keratoconus using Scheimpflug and OCT imaging modalities. Eye Vis 2020;7:48.

33. Scarcelli G, Besner S, Pineda R, Yun SH. Biomechanical characterization of keratoconus corneas ex vivo with Brillouin microscopy. Invest Ophthalmol Vis Sci 2014;55:4490–4495.

34. Elharrouss O, Akbari Y, Almaadeed N. Backbonesreview: Feature extraction networks for deep learning and deep reinforcement learning approaches. ArXiv; 2022. https://doi.org/https://doi.org/10.48550./arXiv.2206.08016

35. Ferreira-Mendes J, Lopes BT, Faria-Correia F, Salomão MQ, Rodrigues-Barros S, Ambrósio R Jr. Enhanced ectasia detection using corneal tomography and biomechanics. Am J Ophthalmol 2019;197:7–16.

36. Kataria P, Padmanabhan P, Gopalakrishnan A, Padmanaban V, Mahadik S, Ambrósio R Jr. Accuracy of Scheimpflug-derived corneal biomechanical and tomographic indices for detecting subclinical and mild keratectasia in a South Asian population. J Cataract Refract Surg 2019;45:328–336.

37. Wu Y, Guo LL, Tian L, Xu ZQ, Li Q, Hu J, et al. Comparative analysis of the morphological and biomechanical properties of normal cornea and keratoconus at different stages. Int Ophthalmol 2021;41:3699–3711.

38. Herber R, Ramm L, Spoerl E, Raiskup F, Pillunat L, Terai N. Assessment of corneal biomechanical parameters in healthy and keratoconic eyes using dynamic bidirectional applanation device and dynamic Scheimpflug analyzer. J Cataract Refract Surg 2019;45:778–788.

39. Li Z, You C, Bhojanapalli S, Li D, Rawat AS, Reddi SJ, et al. Large models are parsimonious learners: Activation sparsity in trained Transformers. ArXiv; 2022. https://doi.org/https://doi.org/10.48550./arXiv.2210.06313

40. Jaszczur S, Chowdhery A, Mohiuddin A, Kaiser Ł, Gajewski W, Michalewski H, et al. Sparse is enough in scaling Transformers. ArXiv; 2021. https://doi.org/https://doi.org/10.48550./arXiv.2111.12763.\

41. Tan Z, Chen X, Li K, Liu Y, Cao H, Li J, et al. Artificial intelligence-based diagnostic model for detecting keratoconus using videos of corneal force deformation. Transl Vis Sci Technol 2022;11:32.

42. Wang W, Duan L, Wang Y. Fast image segmentation using two-dimensional Otsu based on estimation of distribution algorithm. J Electr Comput Eng 2017;2017:1735176.

43. Shorten C, Khoshgoftaar TM. A survey on image data augmentation for deep learning. J Big Data 2019;6:1–48.