Skip to main content
. Author manuscript; available in PMC: 2023 Jun 1.
Published in final edited form as: IEEE Trans Med Imaging. 2022 Jun 1;41(6):1383–1399. doi: 10.1109/TMI.2022.3142610

Fig. 15.

Fig. 15.

Generalized reconstructions for Subjects 2 and 3. The generative network was trained for Subject 1 and applied to Subjects 2 and 3. The HR images reconstructed by SSGNN, as shown in Fig. 14, were used as the gold standards. The generalization for Subject 2 was successful and led to small errors (right column), while failed for Subject 3 with big errors due to the big difference in intensity distributions between Subjects 1 and 3.