Rethinking U-Net from an Attention Perspective with Transformers for Osteosarcoma MRI Image Segmentation

<div>The left side of the figure is the architectural design of the student model/teacher model, and the right side is the overall architecture of Res-Block consisting of convolutional layers, GN, and residual connections.</div>

Computational Intelligence and Neuroscience

fig2

Figure 2

Figure 2: Rethinking U-Net from an Attention Perspective with Transformers for Osteosarcoma MRI Image Segmentation 