EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Glaucoma Classification Using VGG16 with Grad-CAM–Based Explainability

Mahda Laina Arnumukti, Kok Swee Sim

International Journal on Advanced Science, Engineering and Information Technology · 11.10.2026 · 16 (5) · 1522-1527

DOI: 10.18517/ijaseit.16.5.22062 ↗

Kurz gefasst

Fachartikel in der Zeitschrift International Journal on Advanced Science, Engineering and Information Technology, online erschienen am 11.10.2026. Verfasst von Mahda Laina Arnumukti und 1 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Glaucoma is an eye condition that develops gradually. It damages the optic nerve and causes the retina to deteriorate over time. If it is not understood and addressed early, it may result in permanent vision loss. Often referred to as “the invisible thief of sight,” glaucoma may go unnoticed by the patient until it is advanced. Because of this, it is crucial to use medical technology that employs computers to prevent health problems that could become permanent. In this study, digital fundus images are used to automatically analyze and diagnose glaucoma. The dataset used is ACRIMA, which contains retinal images labeled with information from glaucoma patients and healthy individuals, providing data that can be analyzed and trusted for research and assessment. This study employs a learning methodology based on the VGG16 convolutional neural network architecture, which is well known for its strong ability to analyze images and automate classification tasks. Gradient-weighted Class Activation Mapping (Grad-CAM) is used to create visual explanations by highlighting important details in fundus images that directly impact model predictions. The results indicate that the VGG16 model achieved approximately 90% classification accuracy. Grad-CAM integration not only makes the learning process more transparent, but it also increases doctors' confidence in the model's results. This makes it a very useful tool for helping medical professionals make appropriate clinical decisions.

Zitieren

APA

Arnumukti, M. L., & Sim, K. S. (2026). Glaucoma Classification Using VGG16 with Grad-CAM–Based Explainability. International Journal on Advanced Science, Engineering and Information Technology, 16(5), 1522-1527. https://doi.org/10.18517/ijaseit.16.5.22062

Vancouver

Arnumukti ML, Sim KS. Glaucoma Classification Using VGG16 with Grad-CAM–Based Explainability. International Journal on Advanced Science, Engineering and Information Technology. 2026;16(5):1522-1527. doi:10.18517/ijaseit.16.5.22062

BibTeX
@article{Arnumukti2026,
  author = {Arnumukti, Mahda Laina and Sim, Kok Swee},
  title = {Glaucoma Classification Using VGG16 with Grad-CAM–Based Explainability},
  journal = {International Journal on Advanced Science, Engineering and Information Technology},
  year = {2026},
  volume = {16},
  number = {5},
  pages = {1522--1527},
  doi = {10.18517/ijaseit.16.5.22062}
}
Kostenlos starten →