Table 4.
LPBA40 | μ (SD) | IBSR18 | μ (SD) | CUMC12 | μ (SD) | MGH10 | μ (SD) | |
---|---|---|---|---|---|---|---|---|
Rank 1 | ART | .35 (.07) | SPM_D | .50 (.19) | SPM_D | .47 (.17) | SyN | .39 (.06) |
SyN | .34 (.24) | SyN | .40 (.12) | IRTK | .42 (.07) | ART | .36 (.07) | |
IRTK | .35 (.15) | SyN | .41 (.06) | |||||
ART | .33 (.08) | ART | .35 (.05) | |||||
2 | JRD-fluid | .18 (.13) | ||||||
3 | JRD-fluid | .20 (.08) | FNIRT | .06 (.11) | JRD-fluid | .07 (.07) | IRTK | .26 (.07) |
IRTK | .18 (.15) | D. Demons | .01 (.08) | FNIRT | .07 (.09) | SPM_D | .25 (.28) | |
FNIRT | .17 (.08) | ROMEO | .01 (.28) | D. Demons | .05 (.05) | |||
SPM_D | .14 (.31) |
This table lists the methods that attained the top three ranks after averaging scores across all brain regions then across all registration pairs (μ=mean, SD=standard deviation). The scores reflect a pairwise comparison between methods, according to target overlap (see text). Methods within ranks 1, 2, and 3 have positive means lying within one, two, and three standard deviations of the highest mean, respectively. Values are not comparable across label sets (columns). (SPM_D=DARTEL pairwise).