Deep Convolutional Neural Networks-Based Automatic Breast Segmentation and Mass Detection in DCE-MRI

<div>The architecture of U-Net. The network consists of two parts. The left half is feature extraction. After each pool layer, the scale of feature changes. There are five scales in total. The right half is upsampling. Every time the feature is upsampled, it will be fused with the same scale corresponding to the feature extraction part.</div>

Computational and Mathematical Methods in Medicine

fig2

Figure 2

Figure 2: Deep Convolutional Neural Networks-Based Automatic Breast Segmentation and Mass Detection in DCE-MRI 