TY - JOUR
T1 - A method for the detection of diabetic retinopathy, by analyzing the image of the retinal vascular network
AU - Luna-Benoso, B.
AU - Martínez Perales, J. C.
AU - Flores-Carapia, R.
AU - Barrales López, A. L.
PY - 2014
Y1 - 2014
N2 - Diabetic retinopathy is a complication of diabetes caused by changes in the blood vessels of the retina, can cause blindness if not detected and treated early. Exudates and hemorrhages are two major signs present in diabetic retinopathy. This paper presents an automated method for classifying patients suering from diabetic retinopathy by analyzing the image of the retinal vascular network. The proposed method is divided into three stages: segmentation of hemorrhages and exudates by digital image processing of retinal vascular network using spatial domain methods, extraction of characteristics using Hu moments, and pattern recognition by using associative memories based on cellular automata.
AB - Diabetic retinopathy is a complication of diabetes caused by changes in the blood vessels of the retina, can cause blindness if not detected and treated early. Exudates and hemorrhages are two major signs present in diabetic retinopathy. This paper presents an automated method for classifying patients suering from diabetic retinopathy by analyzing the image of the retinal vascular network. The proposed method is divided into three stages: segmentation of hemorrhages and exudates by digital image processing of retinal vascular network using spatial domain methods, extraction of characteristics using Hu moments, and pattern recognition by using associative memories based on cellular automata.
KW - Associative memories
KW - Digital Imaging
KW - Exudates
KW - Hemorrhages
KW - Hu invariant moments
UR - http://www.scopus.com/inward/record.url?scp=84890708557&partnerID=8YFLogxK
U2 - 10.12988/ces.2014.31064
DO - 10.12988/ces.2014.31064
M3 - Artículo
SN - 1313-6569
VL - 7
SP - 117
EP - 134
JO - Contemporary Engineering Sciences
JF - Contemporary Engineering Sciences
IS - 1-4
ER -