Skip to main content
. Author manuscript; available in PMC: 2022 Nov 14.
Published in final edited form as: Biomed Image Regist (2022). 2022 Jul 9;13386:103–115. doi: 10.1007/978-3-031-11203-4_12

Table 2.

Ablation of network and training hyperparameters used and their influence on the best epoch validation accuracy. Default hyperparameter: (6, MSE, False, True).

Number of levels Training loss function Multi-scale loss Multi-scale warp
6 7 Dice EPE MSE True False True False
Dice 0.927 0.947 0.954 0.942 0.947 0.939 0.947 0.947 0.903
EPE 0.450 0.122 0.103 0.195 0.122 0.270 0.122 0.122 0.738