Table 3 Quantitative assessment of fat and water predictions compared with the original data, using all 800 scans from the cross-validation experiments for training and an out-of-sample test set of 227 scans for evaluation

\(IP\rightarrow \hat{F},\hat{W}\)

0.926 \({\pm }\) 0.010

24.77 \({\pm }\) 0.73

0.949 \({\pm }\) 0.008

25.22 \({\pm }\) 0.79

\(IP,OP\rightarrow \hat{F},\hat{W}\)

0.967 \({\pm }\) 0.005

29.47 \({\pm }\) 0.86

0.977 \({\pm }\) 0.004

29.74 \({\pm }\) 0.91

\(IP,OP\rightarrow \hat{F},\hat{W}\)

0.939 \({\pm }\) 0.008

26.48 \({\pm }\) 0.90

0.953 \({\pm }\) 0.007

26.73 \({\pm }\) 0.92

(Dixon generator loss)