Table 1 Data split for training and quantitatively testing automated heart segmentation (initial cohort)

From: Deep learning-based detection of murine congenital heart defects from µCT scans

 

Train

Test

Total

Stage

Normal

CHD

Normal

CHD

 

E18.5

10

9

2

1

22

P0

7

2

4

5

18

Total

28

12

 
  1. Table entries indicate the number of scans with manual segmentations from the initial cohort used for training and testing the nnU-net segmentation framework. CHD Congenital Heart Defect, E18.5, P0: developmental stages at which scans were taken.