Table 4 Cross-modality evaluation: Training on LiTS (CT) and testing on CHAOS (MRI) without fine-tuning

From: Prompt-mamba filtering networks for accurate hepatocellular carcinoma lesion segmentation in abdominal CT

Method

Dice (%)

HD95 (mm)

Recall (%)

Prec. (%)

U-Net

70.5

18.4

72.1

69.2

Attention U-Net

72.0

17.5

73.8

70.6

U-Net++

72.6

17.0

74.3

71.0

U-Net3+

73.2

16.7

74.9

71.6

TransUNet

74.8

15.2

76.2

73.4

Swin-UNet

77.3

14.1

78.0

75.9

MISSFormer

76.9

14.4

77.6

75.3

HiFormer

77.8

13.9

78.7

76.1

MERIT

79.3

13.2

80.1

77.5

MedSAM

80.1

13.0

81.0

77.8

MedSegDiff

78.6

13.6

79.4

76.9

Prompt-Mamba-AF

82.4

12.5

83.6

79.2

  1. The bold means the best results in the table.
Back to article page