Skip to main content
Fig. 3 | Radiation Oncology

Fig. 3

From: Dose prediction for cervical cancer VMAT patients with a full-scale 3D-cGAN-based model and the comparison of different input data on the prediction results

Fig. 3

The proposed encoding-decoding generator architecture improved by UNet3+. \({S}_{i}^{EN}\)and \({S}_{i}^{DE}\) represent the encoding and decoding steps respectively. The size and the number of channels are marked below the feature maps. The orange and blue arrows represent the convolution block and the downsampling blocks, and the straight lines with arrows in the figure represent the full-scale feature connection and fusion operation

Back to article page

Radiation Oncology

ISSN: 1748-717X

Contact us