Authors : Akande Noah Oluwatobi, Abikoye Oluwakemi Christianah, Ayegba Peace, Gbadamosi Babatunde and Adegun Adekanmi Adeyinka
Abstract: Diabetic retinopathy is a micro vascular complication of diabetes which alters the retinal vascular patterns. The increase in this alteration may result in the complete damage of retinal blood vessels and eventually lead to total sight loss or blindness. However, early diagnosis and detection of these changes in the blood vessel patterns could help diagnose Diabetic Retinopathy (DR), determine its stage and define the required treatment to be proffered. This data study presents image files of 50 high resolution blood vessels images that were segmented from Diabetic Retinopathy Image Database (DRIDB). Prior to blood vessel extraction, the green component of the raw images was extracted and converted to a grayscale image. Afterwards, the images were preprocessed using median filter, mahalanobis distance and contrast limited adaptive histogram equalization. The segmented blood vessels can be used to study blood vessel patterns with a view to detect any alteration that could signify the presence of DR. Furthermore, the automatically segmented blood vessels can be compared with other manually segmented blood vessels in order to validate the technique adopted. Prospective researchers can also make use of the blood vessels in developing retina recognition systems.
Akande Noah Oluwatobi, Abikoye Oluwakemi Christianah, Ayegba Peace, Gbadamosi Babatunde and Adegun Adekanmi Adeyinka, 2019. Segmented Retinal Blood Vessels of Healthy and Diabetic Retinopathy Individual. Journal of Engineering and Applied Sciences, 14: 5794-5799.