Integrating prior knowledge into a bibranch pyramid network for medical image segmentation

Xianjun Han, Tiantian Li, Can Bai, Hongyu Yang

Abstract: Highlights•The axial attention mechanism can be more effective in reducing computational pressure.•The bi-branch model is capable of extracting features with greater efficiency.•The multi-scale representation of the pyramid structure makes the segmentation model more robust.•The encoder of SAM improved the segmentation performance.