Table 3 Comparative experiment on mouse membranous nephropathy.

From: Rapid diagnosis of membranous nephropathy based on kidney tissue Raman spectroscopy and deep learning

Model

GFlops

Parameters count

Accuracy

Precision

Sensitivity

Specificity

F1-score

AUC

LDA

96.0

97.6

96.0

94.29

96.17

97.14

DT

88.0

90.5

88.0

89.14

88.45

90.21

LR

96.0

97.17

96.0

97.14

96.2

97.32

RF

92.0

93.17

92.0

93.14

92.2

93.32

LSTM

0.17

0.09 M

82.0

87.29

84.17

84.17

81.39

84.17

DNN

0.002

0.05 M

88.0

91.21

88.33

88.33

87.16

88.33

AlexNet

0.62

3.40 M

86.0

88.95

87.5

87.5

85.72

87.5

ResNet

8.45

15.98 M

82.0

89.05

80.0

80.0

77.94

80.0

DenseNet

5.21

7.43 M

94.0

95.14

93.33

93.33

93.58

93.33

CNN-BiLSTM

1.04

0.26 M

98.0

98.0

98.33

98.33

97.98

98.33