Table 5 Model evaluation using different feature sets for extra limited resources.

From: Early prediction of intraventricular hemorrhage in very low birth weight infants using deep neural networks with attention in low-resource settings

Parameter

Accuracy

Precision

Recall

F1 score

Specificity

FPR

FNR

MCC

Marked-ness

Feature set 1

 Birth weight (g)

0.84

0.88

0.84

0.84

0.68

0.31

0.00

0.72

0.76

 Gestational age

 APGAR score at 1 min

 Sex (male)

Feature set 2

 Delivery mode

0.79

0.82

0.79

0.78

0.63

0.36

0.05

0.61

0.65

 Endotracheal intubation

 Birth weight (g)

 Sex (male)

Feature set 3

 Endotracheal intubation

0.79

0.82

0.79

0.78

0.63

0.36

0.05

0.61

0.65

 Gestational age

 APGAR score at 1 min

 APGAR score at 5 min

Feature set 4

 Maternal age (year)

0.82

0.86

0.82

0.81

0.63

0.36

0.00

0.68

0.74

 Delivery mode

 Endotracheal intubation

 Gestational age