Table 2 Semantic segmentation metrics of UNET variant models with new loss function.

From: A novel sub-differentiable hausdorff loss combined with BCE for MRI brain tumor segmentation using UNet variants

Model

Loss function

Evaluation Metrics

Accuracy

Precision

Recall

Dice_score

IoU

U-Net

SDHL + BCE

99.49

87.58

87.65

85.77

77.54

U-Net+

SDHL + BCE

99.52

87.96

88.03

86.17

77.86

V Net

SDHL + BCE

99.58

88.14

89.68

87.98

78.97

U-Net++

SDHL + BCE

99.67

89.98

90.06

88.59

79.57

Attention U Net

SDHL + BCE

99.71

91.06

91.85

90.16

80.78

Back to article page