Abstract

Optic disc detection and retinal vessel diameter measurement are important tasks in the studies of the association of retinal micro vascular disease with ocular effects, such as cerebral atrophy, cognitive decline, stroke, diabetes and hypertension. This work presents an automatic detection of optic disc irregularities, and determines an appropriate region of interest for retinal vessel diameter measurement. We measured the retinal vessel diameters in 44 diabetic and 60 normal images captured from human eye fundus images with three parameters: Main retinal artery equivalents, Main retinal vein equivalents, and artery to vein diameter ratio. The average of the main artery equivalents in patients with diabetes mellitus is 8.48 pixels, and the average of main vein equivalents is 11.63 pixels. The average of main artery equivalents in healthy people is 9.57 pixels and the average of main vein equivalents is 12.03 pixels. The average ratio of arteriole to vein diameter is about 0.75 in patients’ images with diabetes and approximately 0.80 in healthy people. The results show that the average ratio of arteriole to vein diameter in diabetic patients is smaller than that of healthy people.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call