DocumentCode
1588496
Title
Analysis of Retinal Images Associated with Hypertension and Diabetes
Author
Wang, Xuemin ; Cao, Hongbao ; Zhang, Jie
Author_Institution
Dept. of Biomed. Eng., Tianjin Univ.
fYear
2006
Firstpage
6407
Lastpage
6410
Abstract
Twenty four retinal images from three groups, normotensive, hypertension and diabetes, were analyzed. The features were extracted, including exudates, bifurcation angle, artery-to-veins diameter ratio, mean artery and veins diameters, form and size of optic disc, and vessel tortuosity. The results showed that exudates appeared in both hypertensive and diabetic groups. For hypertensive group, exudates increased with the years of illness. There was no relationship between exudates and years of illness for diabetic group. No significant changes in the first bifurcation angle among three groups were observed. The mean arterial diameter tended to decrease for hypertensive and diabetic groups while the mean veinal diameter dilated. Compared to the normotensive group, the ratio of arterial diameter to veinal diameter decreased for both hypertensive and diabetic group. The form and size changes of the optic disc were obvious in diabetic group. Hypertensive group showed tortuous blood vessels. No tortuous blood vessels were observed in the diabetic group. The study demonstrated that it is feasible to use automatically generated retinal image features as an effective screening method for detection of hypertension and diabetes
Keywords
blood vessels; diseases; eye; feature extraction; medical image processing; artery-to-veins diameter ratio; bifurcation angle; diabetes; exudates; feature extraction; hypertension; normotensive; optic disc; retinal images; tortuous blood vessels; Arteries; Bifurcation; Biomedical imaging; Blood vessels; Diabetes; Feature extraction; Hypertension; Image analysis; Retina; Veins;
fLanguage
English
Publisher
ieee
Conference_Titel
Engineering in Medicine and Biology Society, 2005. IEEE-EMBS 2005. 27th Annual International Conference of the
Conference_Location
Shanghai
Print_ISBN
0-7803-8741-4
Type
conf
DOI
10.1109/IEMBS.2005.1615964
Filename
1615964
Link To Document