Skip to main content
Visual Computing for Industry, Biomedicine, and Art
Fig. 3 | Visual Computing for Industry, Biomedicine, and Art

Fig. 3

From: Two fully automated data-driven 3D whole-breast segmentation strategies in MRI for MR-based breast density using image registration and U-Net with a focus on reproducibility

Fig. 3

Architecture of 3D U-Net for deep learning breast segmentation. The input contains two channels, one for water-only image and the other for fat-only image. The output is segmentation map. The blue boxes represent feature maps. The white boxes are copied feature maps. The number of channels is labeled on the top of the box. Different arrows denote different operations that are indicated in the legends

Back to article page