Data on Retinal Images of Hypertensive Retinopathy and Glaucoma Patients
Description
This dataset contains 110 blood vessels extracted from Digital Retinal Images for Optic Nerve Segmentation Database (DRIONS-DB) publicly available database. It could be used to diagnose several ophthalmic diseases such as Hypertensive retinopathy, Diabetic Retinopathy, Glaucoma etc.
Files
Steps to reproduce
A software was developed by the author to automatically extract blood vessels from Retina images. The blood vessels detection and extraction process involves different stages such as preprocessing and segmentation. Preprocessing was achieved using Median filter, Mahalanobis distance and CLAHE while the segmentation was achieved using Dempster–Shafer (D-S) edge based detector. D-S uses probability-based fusion to merge the outputs of Laplacian of Gaussian (LoG) and canny edge detection filters in determining the continuous paths of a vessel after the starting point has been determined