Unsupervised Domain Adaptation Fundus Image Segmentation via Multi-Scale Adaptive Adversarial Learning

Unsupervised Domain Adaptation Fundus Image Segmentation via Multi-Scale Adaptive Adversarial Learning

IEEE Journal of Biomedical and Health Informatics

10.1109/JBHI.2023.3342422

http://dx.doi.org/10.1109/jbhi.2023.3342422

Wei Zhou, Jianhang Ji, Wei Cui, Yingyuan Wang, Yugen Yi

13 December 2023

Zhou, W., Ji, J., Cui, W., Wang, Y., & Yi, Y. (2024). Unsupervised Domain Adaptation Fundus Image Segmentation via Multi-Scale Adaptive Adversarial Learning. IEEE Journal of Biomedical and Health Informatics, 28(10), 5792–5803. https://doi.org/10.1109/jbhi.2023.3342422

Segmentation of the Optic Disc (OD) and Optic Cup (OC) is crucial for the early detection and treatment of glaucoma. Despite the strides made in deep neural networks, incorporating trained segmentation models for clinical application remains challenging due to domain shifts arising from disparities in fundus images across different healthcare institutions. To tackle this challenge, this study introduces an innovative unsupervised domain adaptation technique called Multi-scale Adaptive Adversarial Learning (MAAL), which consists of three key components. The Multi-scale Wasserstein Patch Discriminator (MWPD) module is designed to extract domain-specific features at multiple scales, enhancing domain classification performance and offering valuable guidance for the segmentation network. To further enhance model generalizability and explore domain-invariant features, we introduce the Adaptive Weighted Domain Constraint (AWDC) module. During training, this module dynamically assigns varying weights to different scales, allowing the model to adaptively focus on informative features. Furthermore, the Pixel-level Feature Enhancement (PFE) module enhances low-level features extracted at shallow network layers by incorporating refined high-level features. This integration ensures the preservation of domain-invariant information, effectively addressing domain variation and mitigating the loss of global features. Two publicly accessible fundus image databases are employed to demonstrate the effectiveness of our MAAL method in mitigating model degradation and improving segmentation performance. The achieved results outperform current state-of-the-art (SOTA) methods in both OD and OC segmentation.

Publisher Copyright

This research / project is supported by the Jiangxi Province Key Subject Academic and Technical Leader Funding Project - N/A
Grant Reference no. : 20212BCJ23017

This research / project is supported by the National Natural Science Foundation of China - N/A
Grant Reference no. : 62062040, 62102270 and 6204170

This research / project is supported by the Project of Natural Science Foundation of Liaoning Province - N/A
Grant Reference no. : 2023-MS-246

This research / project is supported by the Outstanding Youth Project of Natural Science Foundation of Jiangxi - N/A
Grant Reference no. : 20212ACB212003

© 2023 IEEE.  Personal use of this material is permitted.  Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.

https://oar.a-star.edu.sg/communities-collections/articles/21240

2168-2194
2168-2208

Institute for Infocomm Research