TY - JOUR ID - 142210 TI - An Automatic Optic Disk Segmentation Approach from Retina of Neonates via Attention Based Deep Network JO - International Journal of Engineering JA - IJE LA - en SN - 1025-2495 AU - Abaei Kashan, A. AU - Maghsoudi, A. AU - Shoeibi, N. AU - Heidarzadeh, M. AU - Mirnia, K. AD - Department of Mechanical Engineering, Iran University of Science and Technology, Tehran, Iran AD - Department of Mechanical Engineering, Tehran University of Medical Science, Tehran, Iran AD - Department of Ophthalmology, Mashhad University of Medical Sciences, Mashhad, Iran AD - Department of Pediatrics, Tabriz University of Medical Science, Tabriz, Iran AD - Department of Pediatrics, Tehran University of Medical Science, Tehran, Iran Y1 - 2022 PY - 2022 VL - 35 IS - 4 SP - 715 EP - 724 KW - Image Segmentation KW - convolutional neural network KW - attention mechanism KW - Retinopathy of Premature KW - Optic Disk DO - 10.5829/ije.2022.35.04A.11 N2 - Every year, many newborns lose their sight to retinopathy of prematurity (ROP) worldwide. Despite its high prevalence and adverse consequences, periodic examinations can effectively prevent it. The use of an intelligent system enables physicians to avoid medical mistakes while examining newborns. The optic disk (OD) is an integral part of the retina for grading the severity and progression of ROP. Due to the uneven brightness and lack of a defined OD border, the use of retinal images of infants is very challenging for OD diagnosis. This paper provides an innovative model of OD segmentation based on attention gate. Initially, the images were collected and preprocessed and inputted into a novel deep convolutional neural network consisting of attention in skip connections. The architecture is comprised of a two-stage convolutional network. Different outputs are obtained from two individual branches of the original image and image features in the first stage. The outputs were concatenated to transfer into the post-processing stage to identify the area related to the OD. The final results based on the Dice coefficient (Dice) and the Intersection-Over-Union (IoU) were 94.22% and 86.1%, respectively. UR - https://www.ije.ir/article_142210.html L1 - https://www.ije.ir/article_142210_5fe3ae6e309bc5f0d49d4e8faf1fdb72.pdf ER -