Segmentation Of Retinal Blood Vessels Using A Novel Fuzzy Logic Algorithm

Loading...
Thumbnail Image
Authors
Yildirim, Burak
Issue Date
2019-03-11
Type
Thesis
Language
en_US
Keywords
Retinal blood vessels , Fuzzy logic
Research Projects
Organizational Units
Journal Issue
Alternative Title
Abstract
In this work, a rule-based method is presented for blood vessel segmentation in digital retinal images. This method can be used in computer analyses of retinal images, e.g., in automated screening for diabetic retinopathy. Diabetic retinopathy is the most common diabetic eye disease and a leading cause of blindness. Diagnosis of diabetic retinopathy at an early stage can be done through the segmentation of the blood vessels of retina. Many studies have been carried out in the last decade in order to obtain accurate blood vessel segmentation in retinal images including supervised and rule-based methods. This method uses eight feature vectors for each pixel. These features are means and medians of intensity values of pixel itself, first and second nearest neighbor at four directions. Features are used in fuzzy logic algorithm as crisp input. The final segmentation is obtained using a thresholding method. The method was tested on the publicly available database DRIVE and its results are compared with distinguished published methods. Our method achieved an average accuracy of 93.82% and an area under the receiver operating characteristic curve of 94.19% for DRIVE database. Our results demonstrated an average sensitivity of 72.28% and a specificity of 97.04%. The calculated sensitivity and specificity values for DRIVE database also state that the proposed segmentation method is effective and robust.
Description
Citation
B. Yildirim, "Segmentation Of Retinal Blood Vessels Using A Novel Fuzzy Logic Algorithm", Masters thesis, Dept. of Engineering, Univ. of Bridgeport, Bridgeport, CT, 2019.
Publisher
License
Journal
Volume
Issue
PubMed ID
DOI
ISSN
EISSN