Skip to main content
. 2022 Dec 14;44(4):1779–1792. doi: 10.1002/hbm.26174

FIGURE 8.

FIGURE 8

(a) The input patches (patch size = 32) of T1w image, T2w image, and corresponding ground truth segmentation label. (b) The entire 16 feature maps of DU‐Net with MixACB and without (w/o) MixACB. The feature maps are the outputs of the last dense block of DU‐Net. The number in the lower right corner of each feature map indicates the sequence number of the feature map.