Table 5.

Vessel segmentation results using pre-trained Attention U-Net, TAU-Net and hybrid model.

Model	Superior mesenteric artery (SMA) segmentation				Superior mesenteric vein (SMV) segmentation
	Dice	Recall	Precision	95% Hausdorff distance	Dice	Recall	Precision	95% Hausdorff distance
Pre-trained Attention U-Net	0.76 ± 0.08	0.79 ± 0.12	0.76 ± 0.09	3.23 ± 0.45	0.68 ± 0.02	0.77 ± 0.06	0.62 ± 0.07	4.79 ± 0.25
Pre-trained TAU-Net	0.80 ± 0.06	0.88 ± 0.07	0.74 ± 0.09	3.19 ± 0.70	0.70 ± 0.09	0.66 ± 0.12	0.76 ± 0.06	3.23 ± 0.65
Pre-trained hybrid model	0.81 ± 0.06	0.87 ± 0.09	0.76 ± 0.07	2.89 ± 0.75	0.73 ± 0.04	0.81 ± 0.10	0.68 ± 0.03	3.45 ± 0.54

Significant values are in bold.