Table 5.

Results (avg of the metric over the 15 images) using application default parameters and those selected by the tuning algorithms

Workflow	Dice							Jaccard
	Default	NM	PRO	GA	GLCCLUSTER	DIRECT	Spearmint	Default	NM	PRO	GA	GLCCLUSTER	DIRECT	Spearmint
Watershed	0.71	0.80	0.80	0.80	0.78	0.81	0.80	0.57	0.67	0.67	0.67	0.64	0.68	0.68
Level Set	0.61	0.75	0.74	0.82	0.61	0.75	0.86	0.50	0.71	0.65	0.70	0.47	0.63	0.79

The best result for each pair segmentation algorithm and metric of interest is highlighted in bold.