Table 3.

Quantitative analysis of predictions from the trained models on test patients.

	UNet (L1)	UNet (VGG)	UNet++ (L1)	Cycle-GAN PSNR
PSNR	20.169±0.196	19.668 ± 0.189	20.080 ± 0.193	18.279 ± 0.156
SSIM	0.809±0.003	0.728 ± 0.003	0.804 ± 0.003	0.783 ± 0.003
MAE	0.043±0.001	0.047 ± 0.001	0.044 ± 0.001	0.057 ± 0.001
MSE	0.011±0.001	0.011±0.001	0.013 ± 0.001	0.016 ± 0.001

The calculated metrics are presented as mean and standard deviation. The pixel intensities outside the body were excluded when deriving these measurements. The best quantitative metrics are shown in bold.