Table 3 Comparative analysis of MobileDANet and existing works (KMC dataset).

From: MobileDANet integrating transfer learning and dynamic attention for classifying multi target histopathology images with explainable AI

Methodology

Acc (%)

Pre (%)

Rec (%)

F1 score (%)

ResNet5029

75.19

74.58

71.34

71.67

IncResV236

71.92

71.97

69.17

68.96

NASNet37

79.82

78.96

76.92

76.82

ShuffleNet38

83.94

83.51

81.57

81.62

BHCNet39

85.76

84.98

84.85

84.18

BreastNet40

84.67

85.54

84.72

84.93

LiverNet41

86.29

85.65

84.67

84.85

ViT30

81.85

81.01

80.24

80.02

CAiT42

88.18

87.25

88.58

88.97

SwinViT43

87.36

88.41

87.49

87.68

MultiscaleViT44

87.36

87.52

87.36

87.24

CoAtNet45

88.19

88.35

88.28

88.49

ConViT46

87.35

87.49

87.64

87.26

HHFA-Net47

85.69

86.87

85.48

86.22

GCViT47

85.58

86.62

85.51

85.97

APFA-Net48

86.62

87.96

86.56

87.13

RCCGNet5

90.14

89.78

89.60

89.06

RCG-Net24

90.62

91.23

90.63

90.92

MobileDANet (proposed model)

90.71

91.29

90.75

90.94