Table 2.

Listing of state of the art methods to automatically segment the pancreas in CT images

Method	Number of evaluated CT data	Needed CT scans per case	Runtime (min)	Accuracy
Shimizu et al. level set [2]	10	1 (noncontrast)	–	32.5 % overlap
Shimizu et al. atlas [3]	20	3 (multiple-phase)	45	57.9 % overlap
Kitasaka et al. [4]	22	4 (multiple-phase)	–	Visual inspection: 12 high, 6 medium, and 4 no overlap
Wolz et al. atlas [5]	100	–	–	49.6 % overlap
Proposed method	40	1 (portal venous)	20.4	61.2 % overlap

Jaccard index was calculated to show the overlap of the segmentation result and the reference standard

“–” unpublished/unavailable information