Cup and Disc Segmentation in Smartphone Handheld Ophthalmoscope Images with a Composite Backbone and Double Decoder Architecture

Glaucoma is a visual disease that affects millions of people, and early diagnosis can prevent total blindness. One way to diagnose the disease is through fundus image examination, which analyzes the optic disc and cup structures. However, screening programs in primary care are costly and unfeasible....

Full description

Saved in:
Bibliographic Details
Main Authors: Thiago Paiva Freire, Geraldo Braz Júnior, João Dallyson Sousa de Almeida, José Ribamar Durand Rodrigues Junior
Format: Article
Language:English
Published: MDPI AG 2025-04-01
Series:Vision
Subjects:
Online Access:https://www.mdpi.com/2411-5150/9/2/32
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849433867929255936
author Thiago Paiva Freire
Geraldo Braz Júnior
João Dallyson Sousa de Almeida
José Ribamar Durand Rodrigues Junior
author_facet Thiago Paiva Freire
Geraldo Braz Júnior
João Dallyson Sousa de Almeida
José Ribamar Durand Rodrigues Junior
author_sort Thiago Paiva Freire
collection DOAJ
description Glaucoma is a visual disease that affects millions of people, and early diagnosis can prevent total blindness. One way to diagnose the disease is through fundus image examination, which analyzes the optic disc and cup structures. However, screening programs in primary care are costly and unfeasible. Neural network models have been used to segment optic nerve structures, assisting physicians in this task and reducing fatigue. This work presents a methodology to enhance morphological biomarkers of the optic disc and cup in images obtained by a smartphone coupled to an ophthalmoscope through a deep neural network, which combines two backbones and a dual decoder approach to improve the segmentation of these structures, as well as a new way to combine the loss weights in the training process. The models obtained were numerically evaluated through Dice and IoU measures. The dice values obtained in the experiments reached a Dice of 95.92% and 85.30% for the optical disc and cup and an IoU of 92.22% and 75.68% for the optical disc and cup, respectively, in the BrG dataset. These findings indicate promising architectures in the fundus image segmentation task.
format Article
id doaj-art-e2f8c1c6ad17402388a2660102e17fa7
institution Kabale University
issn 2411-5150
language English
publishDate 2025-04-01
publisher MDPI AG
record_format Article
series Vision
spelling doaj-art-e2f8c1c6ad17402388a2660102e17fa72025-08-20T03:26:52ZengMDPI AGVision2411-51502025-04-01923210.3390/vision9020032Cup and Disc Segmentation in Smartphone Handheld Ophthalmoscope Images with a Composite Backbone and Double Decoder ArchitectureThiago Paiva Freire0Geraldo Braz Júnior1João Dallyson Sousa de Almeida2José Ribamar Durand Rodrigues Junior3UFMA/Computer Science Department, Universidade Federal do Maranhão, Campus do Bacanga, São Luís 65085-580, BrazilUFMA/Computer Science Department, Universidade Federal do Maranhão, Campus do Bacanga, São Luís 65085-580, BrazilUFMA/Computer Science Department, Universidade Federal do Maranhão, Campus do Bacanga, São Luís 65085-580, BrazilUFMA/Computer Science Department, Universidade Federal do Maranhão, Campus do Bacanga, São Luís 65085-580, BrazilGlaucoma is a visual disease that affects millions of people, and early diagnosis can prevent total blindness. One way to diagnose the disease is through fundus image examination, which analyzes the optic disc and cup structures. However, screening programs in primary care are costly and unfeasible. Neural network models have been used to segment optic nerve structures, assisting physicians in this task and reducing fatigue. This work presents a methodology to enhance morphological biomarkers of the optic disc and cup in images obtained by a smartphone coupled to an ophthalmoscope through a deep neural network, which combines two backbones and a dual decoder approach to improve the segmentation of these structures, as well as a new way to combine the loss weights in the training process. The models obtained were numerically evaluated through Dice and IoU measures. The dice values obtained in the experiments reached a Dice of 95.92% and 85.30% for the optical disc and cup and an IoU of 92.22% and 75.68% for the optical disc and cup, respectively, in the BrG dataset. These findings indicate promising architectures in the fundus image segmentation task.https://www.mdpi.com/2411-5150/9/2/32fundus imagesegmentationU-Netcomposition backbone segmentation
spellingShingle Thiago Paiva Freire
Geraldo Braz Júnior
João Dallyson Sousa de Almeida
José Ribamar Durand Rodrigues Junior
Cup and Disc Segmentation in Smartphone Handheld Ophthalmoscope Images with a Composite Backbone and Double Decoder Architecture
Vision
fundus image
segmentation
U-Net
composition backbone segmentation
title Cup and Disc Segmentation in Smartphone Handheld Ophthalmoscope Images with a Composite Backbone and Double Decoder Architecture
title_full Cup and Disc Segmentation in Smartphone Handheld Ophthalmoscope Images with a Composite Backbone and Double Decoder Architecture
title_fullStr Cup and Disc Segmentation in Smartphone Handheld Ophthalmoscope Images with a Composite Backbone and Double Decoder Architecture
title_full_unstemmed Cup and Disc Segmentation in Smartphone Handheld Ophthalmoscope Images with a Composite Backbone and Double Decoder Architecture
title_short Cup and Disc Segmentation in Smartphone Handheld Ophthalmoscope Images with a Composite Backbone and Double Decoder Architecture
title_sort cup and disc segmentation in smartphone handheld ophthalmoscope images with a composite backbone and double decoder architecture
topic fundus image
segmentation
U-Net
composition backbone segmentation
url https://www.mdpi.com/2411-5150/9/2/32
work_keys_str_mv AT thiagopaivafreire cupanddiscsegmentationinsmartphonehandheldophthalmoscopeimageswithacompositebackboneanddoubledecoderarchitecture
AT geraldobrazjunior cupanddiscsegmentationinsmartphonehandheldophthalmoscopeimageswithacompositebackboneanddoubledecoderarchitecture
AT joaodallysonsousadealmeida cupanddiscsegmentationinsmartphonehandheldophthalmoscopeimageswithacompositebackboneanddoubledecoderarchitecture
AT joseribamardurandrodriguesjunior cupanddiscsegmentationinsmartphonehandheldophthalmoscopeimageswithacompositebackboneanddoubledecoderarchitecture