Boundary-Aware Network With Topological Consistency Constraint for Optic Chiasm Segmentation

Wei Huang, Xin Shu, Zizhou Wang, Chaoyue Chen, Yang Zhang, Lei Zhang, Jianguo Xu

Abstract: The optic chiasm is a structure that is easily compressed by the tumors leading to different degrees of visual field defect and visual disturbance. When the optic chiasm is compressed, the segmentation of the optic chiasm from magnetic resonance imaging images is helpful for prognosis prediction and radiotherapy planning. However, owing to the ambiguity of the optic chiasm boundary and the neglect of the anatomical structure consistency, the performance of the existing methods is limited. In this article, a boundary-aware network (BANet) with a topological consistency constraint is proposed for the automated segmentation of the optic chiasm. The BANet constrained by topology loss leverages the complementary information between the boundary feature and the segmentation feature to effectively improve segmentation performance and topological consistency. To evaluate the effectiveness of the proposed method, a real-world specialized optic chiasm segmentation dataset is constructed. The experimental results demonstrate that the proposed method achieves higher segmentation accuracy compared with the state-of-the-art method.