Authors : Yuv R Pantha

Volume/Issue : Volume 10 - 2025, Issue 1 - January

Google Scholar : https://tinyurl.com/5t5edrhr

Scribd : https://tinyurl.com/mpdc6jk8

DOI : https://doi.org/10.5281/zenodo.14848301

Abstract : Glaucoma, a leading cause of irreversible blindness, is characterized by progressive loss of retinal ganglion cells (Lee & Mackey, 2022). Early detection and intervention are crucial to prevent vision loss, but diagnosing glaucoma remains a challenge due to its heterogeneous nature and varied presentation (Križaj, 2019). One of the key challenges in diagnosing glaucoma is the lack of consensus within the medical community on standardized diagnostic criteria and treatment protocols (Kroese & Burton, 2003). In recent years, artificial intelligence (AI), particularly deep learning (DL) models, has shown promise in improving glaucoma detection by processing large datasets of ocular images and clinical data (Zhang, Tang, Xia, & Cao, 2023). In theory, AI systems could automate and enhance the accuracy of glaucoma diagnosis, reduce the burden on healthcare professionals, and enable earlier interventions that could prevent vision loss. However, variability in diagnostic thresholds and interpretation methods due to the lack of consensus contributes to inconsistencies across clinical practices. This research explores how these discrepancies in diagnosis, particularly using of retinal nerve fiber layer (RNFL) thickness data contributes to the failure of AI-based glaucoma prediction when applied to real world settings. The paper further discusses the ethical, legal, and clinical implications of AI glaucoma models and suggests that standardized diagnostic criteria and improved collaboration among ophthalmologists and AI developers are essential for enhancing the reliability and applicability of AI in glaucoma detection and management.

Keywords : Glaucoma, Artificial Intelligence, Deep Learning, Optical Coherence Tomography (OCT), Lack of Consensus in Glaucoma, Variability in Glaucoma Diagnosis, AI-Based Glaucoma Prediction, Ethical Implications, Legal Implications.

References :

1. Kroese M, Burton H. Primary open angle glaucoma. The need for a consensus case definition Journal of Epidemiology & Community Health 2003; 57:752-754.
2. Zhang, L., Tang, L., Xia, M., & Cao, G. (2023). The application of artificial intelligence in glaucoma diagnosis and prediction. Frontiers in Cell and Developmental Biology, 11, 1173094.
3. Lee, S. S.-Y., & Mackey, D. A. (2022). Glaucoma – Risk factors and current challenges in the diagnosis of a leading cause of visual impairment. Maturitas, 163, 15–22.
4. Križaj D. What is glaucoma? 2019 May 30. In: Kolb H, Fernandez E, Jones B, et al., editors. Webvision: The Organization of the Retina and Visual System [Internet]. Salt Lake City (UT): University of Utah Health Sciences Center; 1995.
5. Wishart, P. K. (2009). Interpretation of the glaucoma “landmark studies.” British Journal of Ophthalmology, 93(5), 561–562.
6. Shen, D., Wu, G., & Suk, H.-I. (2017). Deep learning in medical image analysis. Annual Review of Biomedical Engineering, 19, 221–248.
7. Ghaffar, F. (2024). Diagnostic decision-making variability in glaucoma: Comparative analysis of novice and expert optometrists to inform AI system design. Borealis, 1.
8. Ahmed, S., Khan, Z., Si, F., Mao, A., Pan, I., Yazdi, F., Tsertsvadze, A., Hutnik, C., Moher, D., Tingey, D., Trope, G. E., Damji, K. F., Tarride, J. E., Goeree, R., & Hodge, W. (2016). Summary of glaucoma diagnostic testing accuracy: An evidence-based meta-analysis. Journal of Clinical Medicine Research, 8(9), 641–649.
9. Budenz, D. L., Anderson, D. R., Varma, R., Schuman, J., Cantor, L., Savell, J., Greenfield, D. S., Patella, V. M., Quigley, H. A., & Tielsch, J. (2007). Determinants of normal retinal nerve fiber layer thickness measured by Stratus OCT. Ophthalmology, 114(6), 1046–1052.
10. Oshitari, T., Hanawa, K. & Adachi-Usami, E. Changes of macular and RNFL thicknesses measured by Stratus OCT in patients with early-stage diabetes. Eye 23, 884–889 (2009).
11. Whang, S. E., & Lee, J.-G. (2020). Data collection and quality challenges for deep learning. Proceedings of the VLDB Endowment, 13(12), 3429–3432.
12. Montesinos López, O.A., Montesinos López, A., Crossa, J. (2022). Overfitting, Model Tuning, and Evaluation of Prediction Performance. In: Multivariate Statistical Machine Learning Methods for Genomic Prediction. Springer, Cham.
13. Jakubovitz, D., Giryes, R., Rodrigues, M.R.D. (2019). Generalization Error in Deep Learning. In: Boche, H., Caire, G., Calderbank, R., Kutyniok, G., Mathar, R., Petersen, P. (eds) Compressed Sensing and Its Applications. Applied and Numerical Harmonic Analysis. Birkhäuser, Cham.
14. Johnson, J.M., Khoshgoftaar, T.M. Survey on deep learning with class imbalance. J Big Data 6, 27 (2019). https://doi.org/10.1186/s40537-019-0192-5