Skip to main content
. Author manuscript; available in PMC: 2013 Jan 22.
Published in final edited form as: Neuroimage. 2011 Jul 30;58(4):1051–1059. doi: 10.1016/j.neuroimage.2011.06.080

Table 4.

Paired comparisons of segmentation by ALVIN, FSL FIRST and FreeSurfer for the young adult, Alzheimer’s disease and infant validation dataset. The consistency of the segmentations is measured using the intraclass correlation coefficient and Dice coefficient.

ALVIN vs FSL
FIRST		 ALVIN vs FreeSurfer FSL vs FreeSurfer
Young Adults ICC 0.835 0.988 0.941
Dice Coefficient (SD) 0.869 (0.052) 0.807 (0.052) 0.757 (0.057)
Alzheimer’s Disease ICC N/Aa 0.987 N/Aa
Dice Coefficient (SD) N/Aa 0.901 (0.015) N/Aa
Infants ICC 0.869 0.990 0.900
Dice Coefficient (SD) 0.753 (0.187) 0.772 (0.094) 0.698 (0.181)
a)

Algorithms with a greater than 50% failure rate were not included in this table