Premium
P3‐171: Robustness of automated hippocampal volumetry across accelerated and nonaccelerated T1‐weighted MRI
Author(s) -
Gray Katherine R.,
Bruen Luke,
Hobbs Nicola Z.,
Wolz Robin,
Hill Derek L.
Publication year - 2015
Publication title -
alzheimer's and dementia
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 6.713
H-Index - 118
eISSN - 1552-5279
pISSN - 1552-5260
DOI - 10.1016/j.jalz.2015.06.1542
Subject(s) - hippocampal formation , neuroimaging , magnetic resonance imaging , dementia , medicine , cognitive impairment , segmentation , standard deviation , mri scan , nuclear medicine , computer science , radiology , disease , pathology , artificial intelligence , mathematics , statistics , psychiatry
has better reproducibility at 1.5T than FreeSurfer at 3T (p1⁄40.008) by 43%. The atrophy rate at 3T for the N1⁄4111 subjects, with the interquartile range in brackets, was -0.1% (-2.8, 3.2) for FreeSurfer and -0.1% (-1.4, 1.7) for MAPS-HBSI. At 1.5T it was 0.3% (-2.3, 2.0) and 0.5% (-1.0, 2.0). The right hippocampi had similar results. Conclusions:The reproducibility of hippocampal atrophy rate measurements for MAPS-HBSI at 3T is about 87% better than FreeSurfer at 3T. FreeSurfer at 1.5T is about 43% more reproducible at 3T. The superior performance of MAPS-HBSI at both 3Tand 1.5T may be due toMAPS-HBSI use of partial volume voxel while FreeSurfer uses discrete voxels.