Dual-branch U-Net architecture for retinal lesions segmentation on fundus image
Authors:
- Ming Yin,
- Toufique Ahmed Soomro,
- Fayyaz Ali Jandan,
- Ayoub Fatihi,
- Faisal Bin Ubaid,
- Muhammad Irfan,
- Ahmed J. Afifi,
- Saifur Rahman,
- Sergii Telenyk,
- Grzegorz Nowakowski
Abstract
Deep learning has found widespread application in diabetic retinopathy (DR) screening, primarily for lesion detection. However, this approach encounters challenges such as information loss due to convolutional operations, shape uncertainty, and the high similarity between different lesions types. These factors collectively hinder the accurate segmentation of lesions. In this research paper, we introduce a novel dual-branch U-Net architecture, referred to as Dual-Branch (DB)-U-Net, tailored to address the intricacies of small-scale lesion segmentation. Our approach involves two branches: one employs a U-Net to capture the shared characteristics of lesions, while the other utilizes a modified U-Net, known as U2Net, equipped with two decoders that share a common encoder. U2Net is responsible for generating probability maps for lesion segmentation as well as corresponding boundary segmentation. DB U-Net combines the outputs of U2Net and U-Net as a dual branch, concatenating their segmentation maps to produce the final result. To mitigate the challenge of imbalanced data, we employ the Dice loss as a loss function. We evaluate the effectiveness of our approach on publicly available datasets, including DDR, IDRiD, and E-Ophtha. Our results demonstrate that DB U-Net achieves AUPR values of 0.5254 and 0.7297 for Microaneurysms and soft exudates segmentation, respectively, on the IDRiD dataset. These results outperform other models, highlighting the potential clinical utility of our method in identifying retinal lesions from retinal fundus images.
- Record ID
- CUT2ed80a3aa304409fb45687e87de7206c
- Publication categories
- ;
- Author
- Journal series
- IEEE Access, ISSN 2169-3536
- Issue year
- 2023
- Vol
- 11
- Pages
- 130451-130465
- Other elements of collation
- fot.; rys.; schem.; tab.; Bibliografia (na s.) - 130463-130464; Bibliografia (liczba pozycji) - 49; Oznaczenie streszczenia - Abstr.; Numeracja w czasopiśmie - Vol. 11
- Keywords in English
- lesions, image segmentation, retina, feature extraction, decoding, convolutional neural networks, shape measurement, deep learning, computer aided diagnosis
- ASJC Classification
- ; ;
- DOI
- DOI:10.1109/ACCESS.2023.3333364 Opening in a new tab
- URL
- https://ieeexplore.ieee.org/document/10319421 Opening in a new tab
- Related project
- [E-1/2023] Faculty of Electrical and Computer Engineering, Cracow University of Technology and the Ministry of Science and Higher Education. . Project leader at PK: , ,
- Language
- eng (en) English
- License
- Score (nominal)
- 100
- Score source
- journalList
- Score
- Publication indicators
- Citation count
- 3
- Additional fields
- Indeksowana w: Web of Science, Scopus
- Uniform Resource Identifier
- https://cris.pk.edu.pl/info/article/CUT2ed80a3aa304409fb45687e87de7206c/
- URN
urn:pkr-prod:CUT2ed80a3aa304409fb45687e87de7206c
* presented citation count is obtained through Internet information analysis, and it is close to the number calculated by the Publish or PerishOpening in a new tab system.