From: Artifact-free fat-water separation in Dixon MRI using deep learning
Model | Water | Fat | SIS | |||||
---|---|---|---|---|---|---|---|---|
SSIM | PSNR (dB) | SSIM | PSNR (dB) | Spleen | Kidney | Liver | Iliopsoas | |
\(IP\rightarrow \hat{F},\hat{W}\) | 0.927Â \({\pm }\)Â 0.011 | 24.83Â \({\pm }\)Â 1.00 | 0.950Â \({\pm }\)Â 0.011 | 25.25Â \({\pm }\)Â 1.12 | 0.858 | 0.802 | 0.965 | 0.935 |
\(IP,OP\rightarrow \hat{F},\hat{W}\) | 0.967Â \({\pm }\)Â 0.008 | 29.47Â \({\pm }\)Â 1.57 | 0.976Â \({\pm }\)Â 0.009 | 29.73Â \({\pm }\)Â 1.67 | 0.926 | 0.934 | 0.988 | 0.990 |
\(IP,OP\rightarrow \hat{F},\hat{W}\) | 0.941Â \({\pm }\)Â 0.009 | 26.50Â \({\pm }\)Â 1.42 | 0.959Â \({\pm }\)Â 0.010 | 27.01Â \({\pm }\)Â 1.56 | 0.895 | 0.898 | 0.965 | 0.959 |
(Dixon generator loss) |