Table 3.

The p-values (in decimals) of the two-tailed paired t-test for Dice scores and surface distance in comparison of the different models.

	Dice score			Surface distance
Model	Left breast	Right breast	Heart	Left breast	Right breast	Heart
BibNet single-organ vs multi-organ	0.0096	0.34	0.89	0.042	0.132	0.796
BibNet multi-organ vs 3-level U-Net	0.0078	0.0061	< 0.0001	0.0040	0.022	< 0.0001
BibNet single-organ vs 3-level U-Net	0.0043	0.0057	< 0.0001	0.0020	0.016	< 0.0001

The scores, which are statistically significant, are marked with a gray background.