Table 3 Diagnostic capacity of the algorithms developed according to the technique used.

From: Machine learning for the development of diagnostic models of decompensated heart failure or exacerbation of chronic obstructive pulmonary disease

Machine learning technique

True positive

False negative

True negative

False positive

Sensitivity *

Specificity *

Accuracy *

Random forest

75

18

138

21

78.3

88.8

83.6

Logistic regression

74

19

129

30

80.8

86.3

83.6

Decision tree

72

21

137

22

78.3

85.8

83.1

Naive Bayes

73

20

142

17

75

90.4

83.1

SVM

77

16

129

30

81.7

85

82.3

LGBM

70

23

132

27

73.3

87.5

80.6

Gradient-boosting classifier

64

29

137

22

69.2

88.3

80.3

KNN

52

41

133

26

53.3

84.2

70.8

  1. *These parameters were obtained from the mean of all patients (since not all had the same number of evaluations, the mean does not necessarily correspond to that obtained from the total true positive, false negative, true negative and false positive data available in the entire sample).
Back to article page