Table 3.

Comparison of DeepHarp with other tools regarding robustness. The bold numbers in second and third columns show DeepHarp results. The bold numbers in sixth and seventh columns show highest robustness Dice scores belonging to HippoDeep.

	Test-Retest Set/Absolute Volume Differences in Milliliter				Test-Retest Set/Dice Scores
	Left Hippocampus Mean Volume Difference ± SD	Right Hippocampus Mean Volume Difference ± SD	p-Values vs. DeepHarp Left Hippocampus	p-Values vs. DeepHarp Right Hippocampus	Left Hippocampus Mean Dice ± SD	Right Hippocampus Mean Dice ± SD	p-Values vs. DeepHarp Left Hippocampus	p-Values vs. DeepHarp Right Hippocampus
DeepHarp	0.080 ± 0.057	0.052 ± 0.036	N/A	N/A	0.95 ± 0.01	0.951 ± 0.005	N/A	N/A
HippoDeep	0.056 ± 0.050	0.044 ± 0.030	<0.163	<0.622	0.957 ± 0.008	0.96 ± 0.006	0.029	<0.001
QuickNat	0.035 ± 0.021	0.055 ± 0.040	<0.004	<0.881	0.948 ± 0.007	0.951 ± 0.006	0.278	0.817
FSL	0.174 ± 0.11	0.103 ± 0.085	<0.003	<0.101	0.918 ± 0.017	0.923 ± 0.012	<0.001	<0.001
FreeSurfer	0.141 ± 0.113	0.114 ± 0.082	<0.078	<0.010	0.91 ± 0.01	0.916 ± 0.008	<0.001	<0.001