Interpretable ResNet-50-Based Classification of Glaucomatous and Non-Glaucomatous Retinal Fundus Images: A Reproducible Multi-Dataset Evaluation

Read full text Summarise with VIRAT
Authors: Vaibhav Gandhi, Priyesh P. Gandhi, Ajay Kumar, Ankit D. Oza, Saritha Rathee, Ashish Agrawal
Cited by: 0

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

Interpretable ResNet-50-Based Classification of Glaucomatous and Non-Glaucomatous Retinal Fundus Images: A Reproducible Multi-Dataset Evaluation Retinal fundus images are not redistributed in this repository in compliance with third-party licensing terms. Users must obtain raw images directly from the ACRIMA, ORIGA, and REFUGE data custodians and specify local paths in the notebook. Data Availability Statement The experimental code, preprocessing scripts, model-configuration files, dataset split specifications, evaluation scripts, and permitted trained-model outputs supporting the findings of this study are available through the project repository/Zenodo record: https://doi.org/10.5281/zenodo.19156142 . The retinal fundus images are not redistributed by the authors. The datasets used in this study are subject to the access conditions and licenses of their respective data providers: ACRIMA: Available through the authorized public repository for glaucoma assessment. ORIGA(-light): Access-controlled dataset. Researchers must request access directly from the original data custodian or authorized repository. The dataset is not redistributed through this study. REFUGE: Available through the authorized repository or registration-based access route, subject to the original dataset terms and conditions. The researchers who wish to replicate the research must gain independent and legal access to the datasets and use the preprocessing configuration and split specification outlined in the repository. The repository does not include raw images or limited dataset files unless a specific license for the dataset allows for their redistribution. The reproducibility may thus be subject to the approval of the original providers of the data sets, especially in the case of ORIGA(-light).

Fields of Study & Subjects

Retinal Imaging and Analysis Glaucoma and retinal disorders Ophthalmology and Visual Health Research

Publication details

Journal / source
Zenodo (CERN European Organization for Nuclear Research)
Publication date
22 March 2026
Language
EN