Table 1 Performance Metrics for Internal Validation.

From: A Custom Annotated Dataset for Segmentation of Pulmonary Veins, Arteries, and Airways

Class

Count

Dice

Recall

Precision

BDR

DLR

Airway Lumen

254

0.941 ± 0.022 (0.845–0.968)

0.975 ± 0.017 (0.896–0.992)

0.909 ± 0.030 (0.756–0.953)

0.978 ± 0.017 (0.925–1.000)

0.989 ± 0.006 (0.959–0.997)

Airway Wall

254

0.866 ± 0.046 (0.686–0.911)

0.853 ± 0.055 (0.634–0.910)

0.881 ± 0.040 (0.723–0.932)

—

—

Pulmonary Arteries

254

0.950 ± 0.008 (0.907–0.965)

0.949 ± 0.010 (0.892–0.966)

0.951 ± 0.012 (0.908–0.972)

—

—

Pulmonary Veins

254

0.953 ± 0.010 (0.882–0.968)

0.950 ± 0.015 (0.850–0.973)

0.956 ± 0.012 (0.901–0.979)

—

—

  1. Data are presented as mean ± standard deviation (minimum-maximum).