Skip to main content

Table 3 Performance of the liver segmentation model on the validation set stratified by the AAST grade

From: Automated quantitative assessment of pediatric blunt hepatic trauma by deep learning-based CT volumetry

AAST grade

Dice (%)

Precision (%)

Recall (%)

RVD (%)

I (n = 3)

95.99 ± 0.38

95.05 ± 2.06

96.99 ± 1.38

2.095 ± 3.651

II (n = 20)

94.99 ± 1.43

94.56 ± 2.24

95.46 ± 1.72

1.006 ± 3.060

III (n = 19)

95.13 ± 0.85

94.40 ± 1.69

95.89 ± 1.02

1.616 ± 2.469

IV (n = 8)

93.54 ± 2.51

92.75 ± 4.52

94.52 ± 3.25

2.172 ± 7.106

V (n = 2)

91.86 ± 3.34

90.81 ± 3.05

92.93 ± 3.66

2.328 ± 0.595

Overall (n = 52)

94.75 ± 1.68

94.11 ± 2.63

95.46 ± 1.96

1.522 ± 3.650