Figure 1 | Scientific Reports

Figure 1

From: Unsupervised-learning-based method for chest MRI–CT transformation using structure constrained unsupervised generative attention networks

Figure 1

Outline of proposed U-GAT-IT + MIND process (G, D, and \(\upeta \) denote generator, discriminator, and auxiliary classifier, respectively). We introduce Cycle loss, which is a comparison within the same domain after two rounds of transformation; this is in addition to MIND loss, which is a comparison between different domains after one round of transformation.

Back to article page