SKA_{pub} | SKA_{opt} | ESKA | SKA_{pub} with unmetabolized ^{18}F-FDG^{*} | SKA_{opt} with unmetabolized ^{18}F-FDG^{*} | ESKA without unmetabolized ^{18}F-FDG^{*} | |||||||

Group | Mean ± SD | r^{2} | Mean ± SD | r^{2} | Mean ± SD | r^{2} | Mean ± SD | r^{2} | Mean ± SD | r^{2} | Mean ± SD | r^{2} |

20 s/55 m (n = 65) | −12% ± 33% | 0.82^{†} | 33^{‡} ± 36%^{§} | 0.93^{†} | −7% ± 22% | 0.98^{†} | −40% ± 24% | 0.86 | −9% ± 22% | 0.97 | 36% ± 33% | 0.98 |

20 s/120 m (n = 8) | −37^{‡} ± 19%^{§} | 0.93 | −36^{‡} ± 20%^{§} | 0.93 | −4^{‡} ± 6%^{§} | 1.0^{†} | −47% ± 15% | 0.93 | −46% ± 16% | 0.93 | 13% ± 6% | 1.0 |

120 s/55 m (n = 51) | 20^{‡} ± 42%^{§} | 0.94 | 35^{‡} ± 47%^{§} | 0.94 | 1^{‡} ± 29%^{§} | 0.93 | −22% ± 20% | 0.94 | −13% ± 22% | 0.93 | 58% ± 65% | 0.93 |

↵* Results using modifications of 3 methods (SKA

_{pub}and SKA_{opt}accounting for U_{mean}or ESKA with U_{mean}set to 0).↵† Results for approach significantly different from other 2 in terms of correlation (Fisher

*r*-to-Z transformation).↵‡ Results for approach significantly different from those of other 2 in terms of mean (Wilcoxon test).

↵§ Results for approach significantly different from those of other 2 in terms of variance (Friedman test).