Boundary and Entropy-driven Adversarial Learning for Fundus Image Segmentation

• Chi-Wing Fu  |  Kang Li  |  Lequan Yu  |  Pheng-Ann Heng  |  Shujun Wang  |  Xin Yang

• Department of Computer Science and Engineering The Chinese University of Hong Kong Hong Kong SAR China  |  Guangdong Provincial Key Laboratory of Computer Vision and Virtual Reality Technology  |  Shenzhen Institutes of Advanced Technology Chinese Academy of Sciences Shenzhen China

Accurate segmentation of the optic disc (OD) and cup (OC) in fundus images from different datasets is critical for glaucoma disease screening. The cross-domain discrepancy (domain shift) hinders the generalization of deep neural networks to work on different domain datasets. In this work, they presented an unsupervised domain adaptation framework, called Boundary and Entropy-driven Adversarial Learning (BEAL), to improve the OD and OC segmentation performance, especially on the ambiguous boundary regions. In particular, the proposed BEAL framework utilizes the adversarial learning to encourage the boundary prediction and mask probability entropy map (uncertainty map) of the target domain to be similar to the source ones, generating more accurate boundaries and suppressing the high uncertainty predictions of OD and OC segmentation. They evaluated the proposed BEAL framework on two public retinal fundus image datasets (Drishti-GS and RIM-ONE-r3), and the experiment results demonstrate that the method outperforms the state-of-the-art unsupervised domain adaptation methods.

Input variables : Images
Output Variables : Segmented image

Visit Model : github.com

Additional links : arxiv.org