Article Per Year (5 Year)
p-Index From 2021 - 2026
0.23
P-Index
This Author published in this journals
All Journal Indonesian Journal of Artificial Intelligence and Data Mining
Iga Novinda Rantaya
Jember University
Computer Science & IT

Published : 1 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search

 1 
Glaucoma Identification on Retinal Fundus Image Using Random Forest Method Iga Novinda Rantaya
Indonesian Journal of Artificial Intelligence and Data Mining Vol 6, No 1 (2023): Maret 2023
Publisher : Universitas Islam Negeri Sultan Syarif Kasim Riau

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24014/ijaidm.v6i1.18765

Abstract

Glaucoma is a disease caused by a buildup of fluid in the eye that can increase intraocular pressure and cause vision loss. This disease cannot be cured, therefore early detection is very important to prevent total vision loss in sufferers. To reduce the errors of observation and diagnosis from doctors, applied computer vision to detect glaucoma in retinal fundus images. The retinal fundus image is first cropped to remove unnecessary parts, then the color image captured by the fundus camera is converted into a grayscale image. The gray scale image will be extracted using the Gray Level Co-occurrence Matrix (GLCM) method. The extracted features will be processed to create a classification model using the Random Forest method which will determine whether the image identified is normal or glaucoma. A series of experiments were conducted on the comparison of training data and testing data, as well as the number of decision trees. Experiments were also conducted on the size of the image cropping and changes in the value of the distance variable in the GCLM feature extraction process. The results of the experiment on an image size of 720x720 pixels and a distance value of 2, obtained a model with an accuracy of 81%, precision 79% and recall 88% in a data comparison of 80:20, and the number of trees as much as 50. The results show that the more number of decision trees does not increase the number of decision trees. accuracy value significantly.
Co-Authors