Retinal blood vessel width measured on color fundus photographs by image analysis

Abstract

The purpose of this paper was to develop a technique using computerized image analysis to measure the width of retinal arteries and veins on color fundus photographs. The width of the retinal vessel was determined from digitized fundus photographs by programs based on edge detection and boundary tracing. The average vessel width was determined at different distances or eccentricities from the center of the optic disc and using various lengths of vessel segment. Measurements of superior and inferior temporal vessel width for 20 eyes by 2 operators showed that about 75% of the measurements could be obtained without the use of plan points. The average percent coefficient of variation of the measurements was 2.17% for 3 measurements of each vessel. This technique provides a reproducible and objective method for obtaining the following parameters: the edges of the vessel, the length of the vessel segment to be measured, location of the measurement along the vessel and subsequent retinal vessel width. This technique is readily suitable for application to clinical studies particularly clinical trials.

Reliability of vessel diameter measurements with a retinal oximeter

“…The most common methods are based on fundus images. Fundus images can be measured manually [35][36][37], semiautomatically [38][39][40][41][42], or automatically [2,[43][44][45].…”

Laser preconditioning on cranial bone site: Analysis of morphological vascular parameters

“…In this present study, we describe a computer‐based semi‐automatic system for the quantification of superficial bone vascular network parameters that had been treated by laser preconditioning prior to X‐ray radiation. Computer‐assisted quantification of vascular parameters have already demonstrated an acceptable degree of accuracy and precision and are usually applied to vascular morphology assessment 20, 21.…”

Detection of Blood Vessels in Retinal Images

“…Various kinds of edge maps are used for vessel tracking. Wu et al 6 and Meehan et al 7 are interested in vessel width and take different edge detectors for finding a suitable piece of vessel. After localization of the optic disk with pyramidal decomposition, Hausdorff-based template matching and confidence assignment, Gagnon et al 8 used a recursive dual edge tracking based on canny algorithm and connectivity recovering beginning at the optic disk.…”