A two-stage multi-scale attention-based network for weakly supervised cataract fundus image enhancement
Abstract Cataract is a major cause of vision loss and hinders further diagnosis. However, enhancing cataract fundus images remains challenging due to limited paired cataract retinal images and the difficulty of recovering fine details in the retinal images. To mitigate these challenges, we in this p...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2025-07-01
|
| Series: | Scientific Reports |
| Subjects: | |
| Online Access: | https://doi.org/10.1038/s41598-025-12157-6 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849343866836090880 |
|---|---|
| author | Xiaoyong Fang Yue Wang Xiangyu Li Wanshu Fan Dongsheng Zhou |
| author_facet | Xiaoyong Fang Yue Wang Xiangyu Li Wanshu Fan Dongsheng Zhou |
| author_sort | Xiaoyong Fang |
| collection | DOAJ |
| description | Abstract Cataract is a major cause of vision loss and hinders further diagnosis. However, enhancing cataract fundus images remains challenging due to limited paired cataract retinal images and the difficulty of recovering fine details in the retinal images. To mitigate these challenges, we in this paper propose a two-stage multi-scale attention-based network (TSMSA-Net) for weakly supervised cataract fundus image enhancement. In Stage 1, we introduce a real-like cataract fundus image synthesis module, which utilizes domain transformation via CycleGAN to generate realistic paired cataract images from unpaired clear and cataract fundus images, thus alleviating the scarcity of paired training data. In Stage 2, we employ a multi-scale attention-based enhancement module, which incorporates hierarchical attention mechanisms to extract rich, fine-grained features from the degraded images under weak supervision, effectively restoring image details and reducing artifacts. Experiments conducted on the Kaggle and ODIR-5K datasets show that TSMSA-Net outperforms existing state-of-the-art methods for cataract fundus image enhancement, even without paired images, and demonstrates strong generalization ability. Moreover, the enhanced images contribute to improved performance in downstream tasks such as vessel segmentation and disease classification. |
| format | Article |
| id | doaj-art-2c17e8adb61940dcb5db5bc7bcde922c |
| institution | Kabale University |
| issn | 2045-2322 |
| language | English |
| publishDate | 2025-07-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Scientific Reports |
| spelling | doaj-art-2c17e8adb61940dcb5db5bc7bcde922c2025-08-20T03:42:49ZengNature PortfolioScientific Reports2045-23222025-07-0115111210.1038/s41598-025-12157-6A two-stage multi-scale attention-based network for weakly supervised cataract fundus image enhancementXiaoyong Fang0Yue Wang1Xiangyu Li2Wanshu Fan3Dongsheng Zhou4Department, School of Safety and Management Engineering, Hunan Institute of TechnologyNational and Local Joint Engineering Laboratory of Computer Aided Design, School of Software EngineeringNational and Local Joint Engineering Laboratory of Computer Aided Design, School of Software EngineeringNational and Local Joint Engineering Laboratory of Computer Aided Design, School of Software EngineeringNational and Local Joint Engineering Laboratory of Computer Aided Design, School of Software EngineeringAbstract Cataract is a major cause of vision loss and hinders further diagnosis. However, enhancing cataract fundus images remains challenging due to limited paired cataract retinal images and the difficulty of recovering fine details in the retinal images. To mitigate these challenges, we in this paper propose a two-stage multi-scale attention-based network (TSMSA-Net) for weakly supervised cataract fundus image enhancement. In Stage 1, we introduce a real-like cataract fundus image synthesis module, which utilizes domain transformation via CycleGAN to generate realistic paired cataract images from unpaired clear and cataract fundus images, thus alleviating the scarcity of paired training data. In Stage 2, we employ a multi-scale attention-based enhancement module, which incorporates hierarchical attention mechanisms to extract rich, fine-grained features from the degraded images under weak supervision, effectively restoring image details and reducing artifacts. Experiments conducted on the Kaggle and ODIR-5K datasets show that TSMSA-Net outperforms existing state-of-the-art methods for cataract fundus image enhancement, even without paired images, and demonstrates strong generalization ability. Moreover, the enhanced images contribute to improved performance in downstream tasks such as vessel segmentation and disease classification.https://doi.org/10.1038/s41598-025-12157-6Cataract fundus enhancementMulti-scale attentionWeakly supervised learning |
| spellingShingle | Xiaoyong Fang Yue Wang Xiangyu Li Wanshu Fan Dongsheng Zhou A two-stage multi-scale attention-based network for weakly supervised cataract fundus image enhancement Scientific Reports Cataract fundus enhancement Multi-scale attention Weakly supervised learning |
| title | A two-stage multi-scale attention-based network for weakly supervised cataract fundus image enhancement |
| title_full | A two-stage multi-scale attention-based network for weakly supervised cataract fundus image enhancement |
| title_fullStr | A two-stage multi-scale attention-based network for weakly supervised cataract fundus image enhancement |
| title_full_unstemmed | A two-stage multi-scale attention-based network for weakly supervised cataract fundus image enhancement |
| title_short | A two-stage multi-scale attention-based network for weakly supervised cataract fundus image enhancement |
| title_sort | two stage multi scale attention based network for weakly supervised cataract fundus image enhancement |
| topic | Cataract fundus enhancement Multi-scale attention Weakly supervised learning |
| url | https://doi.org/10.1038/s41598-025-12157-6 |
| work_keys_str_mv | AT xiaoyongfang atwostagemultiscaleattentionbasednetworkforweaklysupervisedcataractfundusimageenhancement AT yuewang atwostagemultiscaleattentionbasednetworkforweaklysupervisedcataractfundusimageenhancement AT xiangyuli atwostagemultiscaleattentionbasednetworkforweaklysupervisedcataractfundusimageenhancement AT wanshufan atwostagemultiscaleattentionbasednetworkforweaklysupervisedcataractfundusimageenhancement AT dongshengzhou atwostagemultiscaleattentionbasednetworkforweaklysupervisedcataractfundusimageenhancement AT xiaoyongfang twostagemultiscaleattentionbasednetworkforweaklysupervisedcataractfundusimageenhancement AT yuewang twostagemultiscaleattentionbasednetworkforweaklysupervisedcataractfundusimageenhancement AT xiangyuli twostagemultiscaleattentionbasednetworkforweaklysupervisedcataractfundusimageenhancement AT wanshufan twostagemultiscaleattentionbasednetworkforweaklysupervisedcataractfundusimageenhancement AT dongshengzhou twostagemultiscaleattentionbasednetworkforweaklysupervisedcataractfundusimageenhancement |