Building a Lightweight and Time-EfficientHybrid Model for Choroiditis Detection inFundus Images: A Systematic Approach
International Journal of Image, Graphics and Signal Processing · 08.08.2026 · 18 (4) · 1-24
DOI: 10.5815/ijigsp.2026.04.01 ↗Kurz gefasst
Fachartikel in der Zeitschrift International Journal of Image, Graphics and Signal Processing, erschienen am 08.08.2026. Verfasst von Srishti Raj und 1 weiteren Autor:innen.
Abstract (Originalsprache)
While the use of artificial intelligence in ophthalmology has increased over the past few years, inflammatorydiseases like choroiditisremain widely unexplored. A delayed diagnosis in such cases can cause severe complicationsthatcan lead to visual disability, and hence, automated diagnosis systems can be of great importance. In this study, we presenta systematic, evidence-driven methodology for designing a lightweight choroiditis classifier using a real-world smalldataset with a constraint of low resource availability. This hybrid framework consists of two components. The first featureextraction component extracts features using the MobileNet-V3-Small model. The second component, namely theclassification component, utilises the Support Vector Machine as the binary classifier. This optimal combination wasidentified through systematic comparative experiments. Statistical testing confirms the robustness of the classifierselection. The model gives a cross-validation accuracy of 94% and a held-out test accuracy of 97.06% with a training timeof approximately 3 minutes for the end-to-end pipeline on a carefully collected and previously introduced choroiditisdataset. Being lightweight and computationally efficient, this model is a suitable candidate for future development into areliable computer-aided diagnosis tool that could assist experts in reviewing images, providing telemedicine care, andprioritising patient appointments based on the initial results of the automated systems.
Zitieren
APA
Raj, S., & Keshri, A. K. (2026). Building a Lightweight and Time-EfficientHybrid Model for Choroiditis Detection inFundus Images: A Systematic Approach. International Journal of Image, Graphics and Signal Processing, 18(4), 1-24. https://doi.org/10.5815/ijigsp.2026.04.01
Vancouver
Raj S, Keshri AK. Building a Lightweight and Time-EfficientHybrid Model for Choroiditis Detection inFundus Images: A Systematic Approach. International Journal of Image, Graphics and Signal Processing. 2026;18(4):1-24. doi:10.5815/ijigsp.2026.04.01
BibTeX
@article{Raj2026,
author = {Raj, Srishti and Keshri, Anup Kumar},
title = {Building a Lightweight and Time-EfficientHybrid Model for Choroiditis Detection inFundus Images: A Systematic Approach},
journal = {International Journal of Image, Graphics and Signal Processing},
year = {2026},
volume = {18},
number = {4},
pages = {1--24},
doi = {10.5815/ijigsp.2026.04.01}
}
EUVIMED