Skip to main content
. 2021 Aug 31;2021:7552185. doi: 10.1155/2021/7552185

Table 2.

Comparing the segmentation performance of the WBCE-Tversky loss under different hyperparameter configurations.

Loss functions Weights Metrics (%)
DSC F2 PRE RE FPR
WBCE only None 46.7 43.1 62.3 41.6 0.08
Tversky only α=0.50, β=0.50 49.9 46.4 64.3 45.0 0.06
WBCE-Tversky 51.5 49.5 63.2 49.5 0.10
Tversky only α=0.40, β=0.60 51.1 52.5 58.0 51.1 0.14
WBCE-Tversky 52.1 51.5 59.6 52.0 0.10
Tversky only α=0.30, β=0.70 51.5 52.6 59.5 54.8 0.14
WBCE-Tversky 51.9 50.4 62.2 50.3 0.10
Tversky only α=0.20, β=0.80 52.7 55.4 56.7 58.3 0.16
WBCE-Tversky 53.2 55.6 62.6 56.2 0.12
Tversky only α=0.10, β=0.90 50.2 52.7 53.2 56.5 0.20
WBCE-Tversky 51.5 51.6 57.7 53.1 0.14