Table 2.
Mean and standard deviation of the LDDMM energy for four image-to-image test datasets.
| LDDMM energy for image-to-image test datasets | |||
|---|---|---|---|
| LPBA40 | |||
| initial | LO | LP | LPC |
| 0.120 ± 0.013 | 0.027 ± 0.004 | 0.036 ± 0.005 | 0.030 ± 0.005 |
| IBSR18 | |||
| initial | LO | LP | LPC |
| 0.214 ± 0.032 | 0.037 ± 0.008 | 0.058 ± 0.013 | 0.047 ± 0.011 |
| CUMC12 | |||
| initial | LO | LP | LPC |
| 0.246 ± 0.015 | 0.044 ± 0.003 | 0.071 ± 0.004 | 0.056 ± 0.004 |
| MGH10 | |||
| initial | LO | LP | LPC |
| 0.217 ± 0.012 | 0.039 ± 0.003 | 0.062 ± 0.004 | 0.049 ± 0.003 |
initial: the initial LDDMM energy between the original moving image and the target image after affine registration to the atlas space, i.e. the original image matching energy. LO: LDDMM optimization. LP: prediction network. LPC: prediction+correction network.