Possibly conflicting Evaluation Metrics

The metrics used in the score function used for ranking may be conflicting. Namely,

• NAE_ac : lower is better
• DSC : higher is better
• WFSS: higher is better

hence the composite score: Score = 0.5NAE + 0.25DSC + 0.25*WFSS may not rank better algorithms higher.

Thank you for pointing this out. You are completely right. I have updated the evaluation code to reflect this:

score = 0.5 (1 - NAE_AC) + 0.25 DSC + 0.25 WFSS

I am rerunning algorithm evaluations with this updated formula.