AI-based Retinal Image Classification for Early Detection of Eye Diseases
Journal of Soft Computing Paradigm · 26.05.2026 · 8 (2) · 143-157
DOI: 10.36548/jscp.2026.2.004 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Soft Computing Paradigm, online erschienen am 26.05.2026. Verfasst von Ananthakumari A. und 2 weiteren Autor:innen. Die Arbeit nennt 20 Literaturangaben.
Abstract (Originalsprache)
Early detection of eye diseases are essential to prevent severe vision impairment and improve patient care. Eye diseases such as cataract, glaucoma, and diabetic retinopathy are among the leading causes of vision loss if not diagnosed at an early stage. The goal of this project is to use artificial intelligence to develop an automated classification system for retinal images, which uses retinal fundus images to classify the four most common retinal diseases (i.e., cataract, diabetic retinopathy, glaucoma, and normal). We designed an image classification system which uses the Vision Transformer (ViT) deep learning model to classify images into four categories: cataract, diabetic retinopathy, glaucoma, and normal. Before training occurred, all images were pre-processed by using various methods of resizing, normalizing, and applying data augmentations to enhance the model's performance and generalization. The performance metrics used to evaluate the overall performance of the trained model were accuracy, precision, recall, F1-score, confusion matrix, and ROC curve. Overall, the proposed model achieved an average overall accuracy of 93.48% on the test dataset.
Zitieren
APA
A., A., N., M. M., & M., K. (2026). AI-based Retinal Image Classification for Early Detection of Eye Diseases. Journal of Soft Computing Paradigm, 8(2), 143-157. https://doi.org/10.36548/jscp.2026.2.004
Vancouver
A. A, N. MM, M. K. AI-based Retinal Image Classification for Early Detection of Eye Diseases. Journal of Soft Computing Paradigm. 2026;8(2):143-157. doi:10.36548/jscp.2026.2.004
BibTeX
@article{A2026,
author = {A., Ananthakumari and N., Muthu Menaka and M., Kaniska},
title = {AI-based Retinal Image Classification for Early Detection of Eye Diseases},
journal = {Journal of Soft Computing Paradigm},
year = {2026},
volume = {8},
number = {2},
pages = {143--157},
doi = {10.36548/jscp.2026.2.004}
}
EUVIMED