Research Article
A Transfer Deep Generative Adversarial Network Model to Synthetic Brain CT Generation from MR Images
Table 3
PSNR indicators of each model.
| | Sample | FCM | CycleGAN | WGAN | Dual2DGAN | Dual3D&PatchGAN |
| | 1 | 12.4710 | 24.4511 | 19.9934 | 29.3316 | 28.7841 | | 2 | 9.7536 | 27.6829 | 19.5638 | 25.4528 | 29.5963 | | 3 | 13.5412 | 27.3729 | 21.5966 | 26.3986 | 31.7143 | | 4 | 14.6819 | 25.4410 | 22.4536 | 23.5241 | 32.4196 | | 5 | 12.2365 | 22.3596 | 27.5682 | 27.3351 | 27.5655 | | 6 | 14.6512 | 28.5151 | 22.5510 | 29.6637 | 29.5122 | | 7 | 17.7714 | 28.5974 | 22.5809 | 23.4510 | 33.4243 | | 8 | 14.2287 | 24.2586 | 23.4814 | 27.5968 | 29.5399 | | 9 | 15.6632 | 28.6624 | 21.1940 | 28.5129 | 30.0019 | | Means | 13.8887 | 26.3712 | 22.3314 | 26.8074 | 30.2842 | | Variance | 5.1887 | 5.3292 | 5.4789 | 5.2985 | 3.4762 |
|
|
