Abstract
Breast cancer lesion segmentation in DCE-MRI remains challenging due toheterogeneous tumor morphology and indistinct boundaries. To address thesechallenges, this study proposes a novel hybrid segmentation network, HCMA-UNet,for lesion segmentation of breast cancer. Our network consists of a lightweightCNN backbone and a Multi-view Axial Self-Attention Mamba (MISM) module. TheMISM module integrates Visual State Space Block (VSSB) and Axial Self-Attention(ASA) mechanism, effectively reducing parameters through Asymmetric SplitChannel (ASC) strategy to achieve efficient tri-directional feature extraction.Our lightweight model achieves superior performance with 2.87M parameters and126.44 GFLOPs. A Feature-guided Region-aware loss function (FRLoss) is proposedto enhance segmentation accuracy. Extensive experiments on one private and twopublic DCE-MRI breast cancer datasets demonstrate that our approach achievesstate-of-the-art performance while maintaining computational efficiency. FRLossalso exhibits good cross-architecture generalization capabilities. The sourcecode is available at https://github.com/Haoxuanli-Thu/HCMA-UNet.