Skip to main content

Table 2 The performances of the PE models in the validation set of 36 EPE, 46 LPE, 82 Preterm PE and 217 healthy participants

From: Predicting preeclampsia in early pregnancy using clinical and laboratory data via machine learning model

 

Group

True positive

False positive

True negative

False negative

Sensitivity

Specificity

PPV

NPV

Cutoff

Clinical factors

EPE

21

32

185

15

58.33% (95% CI, 42.23%-74.44%)

85.25% (95% CI,80.54%-90%)

39.62% (95% CI,26.45%-52.79%)

92.5% (95% CI,88.85%-96.15%)

0.1

Preterm PE

50

32

185

32

60.98% (95% CI, 50.42%-71.53%)

85.25% (95% CI,80.54%-90%)

60.98% (95% CI,50.42%-71.53%)

85.25% (95% CI,80.54%-90%)

0.17

LPE

30

32

185

16

65.22% (95% CI, 51.45%-78.98%)

85.25% (95% CI, 80.54%-89.97%)

48.39% (95% CI,35.95%-60.83%)

92.04% (95% CI,88.3%-95.78%)

0.35

Clinical factors + lab variables

EPE

26

32

185

10

72.22% (95% CI, 57.59%-86.85%)

85.25% (95% CI, 80.54%-89.97%)

44.83% (95% CI,32.03%-57.63%)

94.87% (95% CI,91.78%-97.97%)

0.09

Preterm PE

52

32

185

30

63.41% (95% CI, 52.99%-73.84%)

85.25% (95% CI, 80.54%-89.97%)

61.9% (95% CI,51.52%-72.29%)

86.05% (95% CI, 81.41%-90.68%)

0.18

LPE

32

32

185

14

69.57% (95% CI, 56.27%-82.86%)

85.25% (95% CI, 80.54%-89.97%)

50% (95% CI,37.75%-62.25%)

92.96% (95% CI,89.41%-96.52%)

0.36