A DATA MINING FRAMEWORK FOR IDENTIFYING IMPORTANT RETINAL FEATURES ASSOCIATED WITH GLAUCOMA
Journal of Digital Security and Forensics · 12.08.2026 · 3 (2) · 20-26
DOI: 10.29121/digisecforensics.v3.i2.2026.117 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Digital Security and Forensics, online erschienen am 12.08.2026. Verfasst von G. Hridya und 1 weiteren Autor:innen. Die Arbeit nennt 19 Literaturangaben.
Abstract (Originalsprache)
The paper "A Data Mining Framework for Identifying Important Retinal Features Associated with Glaucoma" proposes a complete computational method for early detection of glaucoma by systematic analysis of retinal fundus images. The framework comprises image acquisition, pre-processing, segmentation and feature extraction of optic nerve head structures and highlights clinically relevant parameters such as the cup-to-disc ratio and neuroretinal rim thickness. Various machine learning algorithms such as support vector machines, decision tree models, and neural networks are employed for disease presence and severity classification with high sensitivity and specificity than traditional diagnostic methods. This study emphasizes the value of data mining in healthcare, which enables to find slight structural variations associated to glaucoma and to improve the reproducibility of diagnosis. The framework bridges ophthalmic imaging with advanced computational intelligence to provide a scalable and interpretable decision support system for clinicians. The literature review covers existing studies on multimodal imaging, deep feature fusion and explainable AI, tracing the progression from manually designed features to deep learning based models. The framework proposed addresses the existing challenges such as variability in retinal morphology and the limited integration of multimodal data. Future directions include application of deep learning for richer feature representation, application of federated learning for privacy preserving collaboration, and real-time diagnostic applications validated across different populations. To summarize, this study contributes to the field of intelligent, data-driven glaucoma care by offering a unified, reproducible and clinically relevant data mining framework to enhance early detection and enable personalized ophthalmic diagnostics.
Zitieren
APA
Hridya, G., & Chandni, P. (2026). A DATA MINING FRAMEWORK FOR IDENTIFYING IMPORTANT RETINAL FEATURES ASSOCIATED WITH GLAUCOMA. Journal of Digital Security and Forensics, 3(2), 20-26. https://doi.org/10.29121/digisecforensics.v3.i2.2026.117
Vancouver
Hridya G, Chandni P. A DATA MINING FRAMEWORK FOR IDENTIFYING IMPORTANT RETINAL FEATURES ASSOCIATED WITH GLAUCOMA. Journal of Digital Security and Forensics. 2026;3(2):20-26. doi:10.29121/digisecforensics.v3.i2.2026.117
BibTeX
@article{Hridya2026,
author = {Hridya, G. and Chandni, P.M.},
title = {A DATA MINING FRAMEWORK FOR IDENTIFYING IMPORTANT RETINAL FEATURES ASSOCIATED WITH GLAUCOMA},
journal = {Journal of Digital Security and Forensics},
year = {2026},
volume = {3},
number = {2},
pages = {20--26},
doi = {10.29121/digisecforensics.v3.i2.2026.117}
}
EUVIMED