Figure 2
From: Automatic MRI segmentation of pectoralis major muscle using deep learning
The architecture of the three-dimensional convolutional neural network (3D CNN) model used for PM segmentation. Input is a 3D axial T1-weighted PM MRI, followed by a 3 × 3 × 3 3D convolutional layer with 32 filters. Each green block is a ResNet block with Group Normalization. The output of the decoder has the same spatial size as the input, followed by a softmax function.
