Open AccessAutomatic Diabetic Detection System based on Retina Using an artificial intelligent system
Author(s) -
Zayd Assyarif Alaydrus,
Winda Astuti,
Tan Shi
Publication year - 2020
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/982/1/012002
Subject(s) - retina , adaptive histogram equalization , computer science , artificial intelligence , artificial neural network , diabetes mellitus , sample (material) , blood sugar , computer vision , medicine , pattern recognition (psychology) , histogram , histogram equalization , image (mathematics) , biology , neuroscience , chemistry , chromatography , endocrinology
The Diabetes measurement level usually uses the blood sugar test method, where patients have to fast for eight hours before the test. Glucose tests generally only require a small sample of blood using a sterile needle from the glucometer. The sufferer may feel pain due to a pin prick. However, not all diabetics can do glucose tests regarding having a great fear of syringes, especially in older patients. Therefore, in this work, the diabetic detection system with the retina is an alternative for early checking of patients without having to take a sample of the blood patient. The Contrast Limited Adaptive Histogram Equalization (CLAHE) is used to extract the signal, in which the extracted the retina image is later used as input to the artificial neural network (ANN) based identifier. The retina image parameters are compared and classified to identify the human retina that are intended to be performed. The results of the computer simulation show that this technique produces good accuracy 100% and 96.6%, for training and testing phases, respectively.