Skip to main content

Table 2 Performances of the four algorithmic models and pfADA for diagnosing TPE

From: Identifying tuberculous pleural effusion using artificial intelligence machine learning algorithms

 

AUC

Sensitivity

Specificity

PPV

NPV

PLR

NLR

Accuracy

pfADA

0.890

85.4%

84.1%

80.4%

88.3%

5.37

0.17

84.7%

Logistic regression

0.876

80.5%

84.8%

80.2%

85.2%

5.47

0.23

82.9%

KNN

0.895

78.6%

86.6%

82.3%

84.0%

6.28

0.24

83.2%

SVM

0.918

83.2%

85.9%

82.3%

86.6%

6.23

0.20

80.4%

RF

0.971

89.1%

93.6%

91.3%

91.5%

14.97

0.12

91.6%

  1. TPE tuberculous pleural effusion, pfADA pleural fluid adenosine deaminase, KNN k-nearest neighbor, SVM support vector machine, RF random forest, AUC area under the curve, PPV positive predictive value, NPV negative predictive value, PLR positive likelihood ratio, NLR negative likelihood ratio