Glaucoma detection from retinal fundus images using graph convolution based multi-task model
Glaucoma is an abnormality in the eye condition that, if not treated within a safe time limit, can result in visual loss. Glaucoma diagnosis requires professionals to identify minor structural changes in the structure of the optic disc and optic cup from retinal fundus images in a short period. Deep...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2025-03-01
|
| Series: | e-Prime: Advances in Electrical Engineering, Electronics and Energy |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2772671125000385 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850025544905654272 |
|---|---|
| author | Satyabrata Lenka Zefree Lazarus Mayaluri Ganapati Panda |
| author_facet | Satyabrata Lenka Zefree Lazarus Mayaluri Ganapati Panda |
| author_sort | Satyabrata Lenka |
| collection | DOAJ |
| description | Glaucoma is an abnormality in the eye condition that, if not treated within a safe time limit, can result in visual loss. Glaucoma diagnosis requires professionals to identify minor structural changes in the structure of the optic disc and optic cup from retinal fundus images in a short period. Deep learning algorithms have been employed effectively in the segmentation of biomedical images and the identification of diseases. To accomplish good generalization, model training requires comprehensive annotations, which is a difficult task. The intended objective of the present study is to come up with and train a distinctive multi-task deep learning model for automated fundus image segmentation and classification. The multi-task model learns for the segmentation task of Optic Disc (OD) and Optic Cup (OC) and the classification task for accurate glaucoma detection using both structural and image-based features. The multi-task model proposed a modified U-net architecture in which Mobile-Netv2 is used in the encoder part, Graph Convolution Network (GCN) is used in the decoder part, and an attention module (AM) is used to locate the region of interest (ROI) for better feature extraction. The implementation of this model is done using three fundus image datasets such as ORIGA, REFUGE, and DRISTI-GS. The performance of the proposed multi-task model is compared with some existing methods. It shows maximum accuracy of 97.43 % and AUROC of 0.985 for the glaucoma detection task and high-quality OD and OC segmented images with dice coefficient of 97.95 % and 96.11 % respectively for the segmentation task. |
| format | Article |
| id | doaj-art-fac5da15a2c04c26afbbe1fe81756cd8 |
| institution | DOAJ |
| issn | 2772-6711 |
| language | English |
| publishDate | 2025-03-01 |
| publisher | Elsevier |
| record_format | Article |
| series | e-Prime: Advances in Electrical Engineering, Electronics and Energy |
| spelling | doaj-art-fac5da15a2c04c26afbbe1fe81756cd82025-08-20T03:00:49ZengElseviere-Prime: Advances in Electrical Engineering, Electronics and Energy2772-67112025-03-011110093110.1016/j.prime.2025.100931Glaucoma detection from retinal fundus images using graph convolution based multi-task modelSatyabrata Lenka0Zefree Lazarus Mayaluri1Ganapati Panda2Department of Electrical Engineering, C.V. Raman Global University, Bhubaneswar, 752054, Odisha, IndiaDepartment of Electrical Engineering, C.V. Raman Global University, Bhubaneswar, 752054, Odisha, India; Corresponding author.C.V. Raman Global University, Bhubaneswar, 752054, Odisha, IndiaGlaucoma is an abnormality in the eye condition that, if not treated within a safe time limit, can result in visual loss. Glaucoma diagnosis requires professionals to identify minor structural changes in the structure of the optic disc and optic cup from retinal fundus images in a short period. Deep learning algorithms have been employed effectively in the segmentation of biomedical images and the identification of diseases. To accomplish good generalization, model training requires comprehensive annotations, which is a difficult task. The intended objective of the present study is to come up with and train a distinctive multi-task deep learning model for automated fundus image segmentation and classification. The multi-task model learns for the segmentation task of Optic Disc (OD) and Optic Cup (OC) and the classification task for accurate glaucoma detection using both structural and image-based features. The multi-task model proposed a modified U-net architecture in which Mobile-Netv2 is used in the encoder part, Graph Convolution Network (GCN) is used in the decoder part, and an attention module (AM) is used to locate the region of interest (ROI) for better feature extraction. The implementation of this model is done using three fundus image datasets such as ORIGA, REFUGE, and DRISTI-GS. The performance of the proposed multi-task model is compared with some existing methods. It shows maximum accuracy of 97.43 % and AUROC of 0.985 for the glaucoma detection task and high-quality OD and OC segmented images with dice coefficient of 97.95 % and 96.11 % respectively for the segmentation task.http://www.sciencedirect.com/science/article/pii/S2772671125000385Convolutional neural network (CNN)Fundus imageGlaucomaGraph convolution network (GCN)Intra-Ocular pressureOptic disc |
| spellingShingle | Satyabrata Lenka Zefree Lazarus Mayaluri Ganapati Panda Glaucoma detection from retinal fundus images using graph convolution based multi-task model e-Prime: Advances in Electrical Engineering, Electronics and Energy Convolutional neural network (CNN) Fundus image Glaucoma Graph convolution network (GCN) Intra-Ocular pressure Optic disc |
| title | Glaucoma detection from retinal fundus images using graph convolution based multi-task model |
| title_full | Glaucoma detection from retinal fundus images using graph convolution based multi-task model |
| title_fullStr | Glaucoma detection from retinal fundus images using graph convolution based multi-task model |
| title_full_unstemmed | Glaucoma detection from retinal fundus images using graph convolution based multi-task model |
| title_short | Glaucoma detection from retinal fundus images using graph convolution based multi-task model |
| title_sort | glaucoma detection from retinal fundus images using graph convolution based multi task model |
| topic | Convolutional neural network (CNN) Fundus image Glaucoma Graph convolution network (GCN) Intra-Ocular pressure Optic disc |
| url | http://www.sciencedirect.com/science/article/pii/S2772671125000385 |
| work_keys_str_mv | AT satyabratalenka glaucomadetectionfromretinalfundusimagesusinggraphconvolutionbasedmultitaskmodel AT zefreelazarusmayaluri glaucomadetectionfromretinalfundusimagesusinggraphconvolutionbasedmultitaskmodel AT ganapatipanda glaucomadetectionfromretinalfundusimagesusinggraphconvolutionbasedmultitaskmodel |