Table 4.

Summary of automated segmentation methods of the hippocampus.

No of atlases	DSCmean	Reference study	Validation	Dataset (ground-truth)
9	0.881	AWoL-MRF	Three-Fold MCCV on 60 subjects	ADNI (Pruessner)
9	0.897	AWoL-MRF	Three-Fold MCCV on 81 subjects	FEP (Pruessner)
9	0.807	AWoL-MRF	One-Fold MCCV on 44 subjects	3-step segmentation protocolb
9	0.869	Pipitone et al., 2014	10-Fold MCCV on 60 subjects	ADNI (Pruessner)
9	0.892	Pipitone et al., 2014	Five-Fold MCCV on 81 subjects	FEP subjects
9	0.79	Guo et al., 2015	One-Fold MCCV on 44 subjects	3-step segmentation protocolb
30	0.82	Heckemann et al., 2006	LOOCV	Controls
21	0.862	Morra et al., 2008	LOOCV	ADNI (SNT)
55	0.86	Barnes et al., 2007	LOOCV	Controls and AD
275	0.835	Aljabar et al., 2009	LOOCV	Controls
80	0.89	Collins and Pruessner, 2010	LOOCV	Controls
30	0.885	Lötjönen et al., 2010	Segmentation of 60 subjects	ADNI (SNT)
55	0.89	Leung et al., 2010	Segmentation of 30 subjects	ADNI (SNT)
30	0.848	Wolz et al., 2010	Segmentation of 182 subjects	ADNI (SNT)
16	0.861	Coupé et al., 2011a	LOOCV	ADNI (Pruessner)
20	0.897 (L–HC)	Wang et al., 2012	10-Fold MCCV on 20 of 139 subjects	Landmark based semi-automatic segmentation + manual correction
	0.888 (R–HC)
15	0.862(L–HC)	Wang et al., 2013c	Segmentation of 20 subjects (JLF)	BrainCOLOR
	0.861(R–HC)
15	0.872(L–HC)	Wang et al., 2013c	Segmentation of 20 subjects (With corrective learning)	BrainCOLOR
	0.871(R–HC)
9	0.841	Pipitone et al., 2014	10-Fold MCCV on 69 subjects	ADNI (SNT)

AD, Alzheimer's Disease; MCI, Mild Cognitive Impairment; CN, Cognitively Normal; FEP, First Episode of Psychosis; LOOCV, Leave-one-out cross-validation; MCCV, Monte Carlo cross-validation; SNT, Surgical Medtronic Navigation Technologies semi-automated labels; L–HC, Left hippocampus; R–HC, Right hippocampus.

^aAD: 0.838, MCI: n/a, CN: 0.883.

^bSee Guo et al. (2015) for manual segmentation protocol details.

^cThe method were applied in the 2012 MICCAI Multi-Atlas Labeling Challenge.