Skip to main content

Table 3 Confusion matrices of developed models

From: Application of artificial intelligence in a real-world research for predicting the risk of liver metastasis in T1 colorectal cancer

Confusion matrix

Inner validation

Outer validation

Actual

Prediction

Actual

Prediction

LM (−)

LM (+)

LM (−)

LM (+)

LGBM

LM (+)

42

113

LM (+)

4

4

 

LM (−)

3123

79

LM (−)

317

1

RF

LM (+)

46

109

LM (+)

3

5

 

LM (−)

3136

66

LM (−)

318

0

GNB

LM (+)

32

123

LM (+)

0

8

 

LM (−)

3051

151

LM (−)

313

5

KNN

LM (+)

49

106

LM (+)

4

4

 

LM (−)

3111

91

LM (−)

316

2

MLP

LM (+)

64

91

LM (+)

5

3

 

LM (−)

3131

71

LM (−)

303

15

CART

LM (+)

41

114

LM (+)

3

5

 

LM (−)

3100

102

LM (−)

313

5

SVM

LM (+)

35

120

LM (+)

0

8

 

LM (−)

3059

143

LM (−)

293

25

Stacking

LM (+)

26

129

LM (+)

0

8

 

LM (−)

3062

140

LM (−)

303

15

  1. LM liver metastasis, LGBM Light Gradient Boosting Decision, RF Random Forest, GNB Gaussian Naive Bayesian, KNN K-Nearest Neighbor, MLP Multilayer Perceptron, CART Classification and Regression Trees, SVM Support Vector Machine