Open AccessAutomatic Detection of New Vessels in Retinal Photographs using Machine Learning
Author(s) -
M. Moorthi,
M. Gopikrishnan,
V Pooja,
M R Ragavi Priya,
D K Vasanthakumar
Publication year - 2021
Publication title -
iop conference series. materials science and engineering
Language(s) - English
Resource type - Journals
eISSN - 1757-899X
pISSN - 1757-8981
DOI - 10.1088/1757-899x/1187/1/012030
Subject(s) - optic disc , diabetic retinopathy , support vector machine , artificial intelligence , retinal , retina , computer science , computer vision , image processing , ophthalmology , medicine , image (mathematics) , diabetes mellitus , biology , neuroscience , endocrinology
Diabetic Retinopathy is a rare micro-vascular complication that leads to severe ocular disorder or visual impairment. This complication is unidentified due to the mild progression of its symptoms that is associated with Diabetic Mellitus. The disease is categorized by the development of abnormal new vessels developing the optical disc of the retina of the eye. In this paper, an automatic detection system for detecting new vessels in the optical disc region is detected by using image processing techniques in advance. Segmenting the vessels from the optic disc by extracting the desired region of interest. Based on this, the features are extracted and detected as the presence of new vessels in the optic disc by using a Support Vector Machine (SVM) classifier. This automatic system is trained and tested by using 65 normal images that are categorized as 33 normal images and 32 abnormal images. The discrimination performance of the system shows an accuracy of 95.38% in detecting the new vessels. It obtained sensitivity and specificity of 96.8% and 93.9% respectively.