EUVIMEDEUVIMED entdecken
Open Access · CC BY-SA 4.0

Evaluating the Generalizability of Machine Learning Models for Glaucoma Detection

Nadia Rasool, Mahmood Ali, Amna Manzoor, Tehreem Tanveer, Farah Akhter, Furqan Shaukat

Journal of Rawalpindi Medical College · 30.06.2026 · 30 (2) · 236-244

DOI: 10.37939/jrmc.v30i2.3076 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Rawalpindi Medical College, online erschienen am 30.06.2026. Verfasst von Nadia Rasool und 5 weiteren Autor:innen. Die Arbeit nennt 24 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-SA 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Objective

The proposed research aims to develop a novel deep learning method based on a 2D Convolutional Neural Network (CNN), which would successfully merge information from two essential ocular imaging techniques, including fundus photography and optical coherence tomography.

Methods

The proposed CNN model achieved both high diagnostic precision and generalization capability through its use of the RMSprop optimizer and categorical cross-entropy loss function. Multiple publicly available datasets were used to evaluate the proposed method, including RIM-ONE, Drishti-GS, ORIGA-light, HRF, and ACRIMA (fundus images), as well as a specialized clinical OCT dataset. The proposed method was tested using standard performance metrics against previous notable studies.

Results

The proposed CNN model produced AUC scores of 0.97 for OCT images and 0.96 for fundus images, which outperformed previous results. The OCT images delivered vital information about the optic nerve head structure, which fundus images confirmed through their display of essential clinical indicators for glaucoma diagnosis.

Conclusion

The research shows that the proposed deep learning method can perform real-time glaucoma detection during medical procedures. The proposed model produced exceptional diagnostic results for subclinical glaucoma suspects through its dual imaging system, which improved the ability to protect vision quality and reduce the impact on life quality. Keywords: Glaucoma, Fundus Oculi, Diagnostic Imaging, Computer-Aided Diagnosis, OCT

Zitieren

APA

Rasool, N., Ali, M., Manzoor, A., Tanveer, T., Akhter, F., & Shaukat, F. (2026). Evaluating the Generalizability of Machine Learning Models for Glaucoma Detection. Journal of Rawalpindi Medical College, 30(2), 236-244. https://doi.org/10.37939/jrmc.v30i2.3076

Vancouver

Rasool N, Ali M, Manzoor A, Tanveer T, Akhter F, Shaukat F. Evaluating the Generalizability of Machine Learning Models for Glaucoma Detection. Journal of Rawalpindi Medical College. 2026;30(2):236-244. doi:10.37939/jrmc.v30i2.3076

BibTeX
@article{Rasool2026,
  author = {Rasool, Nadia and Ali, Mahmood and Manzoor, Amna and Tanveer, Tehreem and Akhter, Farah and Shaukat, Furqan},
  title = {Evaluating the Generalizability of Machine Learning Models for Glaucoma Detection},
  journal = {Journal of Rawalpindi Medical College},
  year = {2026},
  volume = {30},
  number = {2},
  pages = {236--244},
  doi = {10.37939/jrmc.v30i2.3076}
}
Kostenlos starten →