Figure 2.
The basic structure of DGCA V-Net. The changes in the number of image channels caused by some operations in the figure have been marked above these operations. DGCA V-Net is an end-to-end network that integrates the GGCA module, the GSA module, and the DGA module into a V-Net model. Conv, convolution; DGA, dual-input-guided feature aggregation; DGCA, dual-channel grouped cross-dimension attention; ELU, exponential linear unit; GGCA, global grouped coordinate attention; GSA, grouped split attention.
