Table 2.

The results of tumor segmentation on the 3DIRCADb data set. The best results are shown in bold.

Model	DC	VOE	RVD	ASSD	HD
Li et al. (2013b)	-	14.40 (±5.30)	-8.10 (±2.10)	2.40 (±0.80)	2.90 (±0.70)
H-DenseUNet (Li et al., 2018)	93.70 (±2.00)	11.68 (±4.33)	-0.01 (±0.05)	0.58 (±0.46)	1.87 (±2.33)
MRFNet (Liu et al., 2020d)	94.81 (±4.20)	6.87 (±5.98)	0.07(±0.16)	0.82 (±0.64)	6.74 (±0.64)
MedT (Valanarasu et al., 2021)	94.99 (±2.43)	6.57 (±4.38)	0.56 (±0.30)	0.73 (±0.58)	4.20 (±1.03)
TransUNet (Chen et al., 2021)	95.06 (±1.89)	6.09 (±3.97)	0.54 (±0.22)	0.69 (±0.74)	2.01(±0.89)
LLRHNet	95.06 (±1.31)	6.04 (±4.67)	0.43 (±0.12)	0.58 (±0.66)	1.93 (±0.71)