Citation

Saad, Assyareefah Hudaibah and Wan Zaki, Wan Mimi Diyana and Mat Daud, Marizuana and Amir Hamzah, Nur Asyiqin and Mustapha, Aouache (2024) Pixel-based Quantification of Retinal Vascular Width using Distance Transform Metric Technique. ICCTA '24: Proceedings of the 2024 10th International Conference on Computer Technology Applications. pp. 149-156.

Text n.pdf - Published Version Restricted to Repository staff only Download (1MB)

Abstract

The retinal vasculature network serves as a crucial indicator of overall human health, given its structural and physiological parallels with various body systems. Quantifying retinal vasculature width is significant for evaluating ocular abnormalities, vascular caliber, and systemic conditions such as hypertension, cardiovascular issues, and nervous system disorders. However, current methodologies face challenges in effectively integrating pixel data for accurate vessel width assessment. This study introduces an optimal quantification approach that incorporates all pixel data without disregarding any connected pixels, utilizing diverse Distance Transform (DT) metrics. Three datasets—Digital Retinal Images for Vessel Extraction (DRIVE), High Resolution Fundus (HRF), and LES-AV—each comprising 20, 45, and 22 fundus images, respectively, were employed. We investigated various DT metrics, including Euclidean DT (EDT), Chessboard DT, Cityblock DT, and Quasi-Euclidean DT (QEDT). Following meticulous metric selection, EDT and QEDT demonstrated promising results that met our research requirements, showing significant statistical analyses (p < 0.05 via Two-sample T-test and Tukey post-hoc). Notably, QEDT emerges as the most suitable and statistically significant metric for vessel quantification, meeting the objective to quantify vascular width with the shortest distance possible while incorporating all connected pixels. This finding underscores QEDT's capacity to quantify retinal blood vessels (RBV), highlighting its potential in advancing ophthalmology diagnostics and treatments.

Item Type:	Article
Uncontrolled Keywords:	Retinal vascular width quantification, Distance transform metric, retinal image processing
Subjects:	Q Science > QA Mathematics
Divisions:	Faculty of Engineering and Technology (FET)
Depositing User:	Ms Nurul Iqtiani Ahmad
Date Deposited:	04 Nov 2024 02:09
Last Modified:	04 Nov 2024 02:09
URII:	http://shdl.mmu.edu.my/id/eprint/13123

Downloads

Edit (login required)