Premium
IC‐P‐027: Robust automatic segmentation of hippocampus differentiates normal aging from Alzheimer's dementia in the ADNI cohort
Author(s) -
Collins D. Louis,
Fonov Vladimir S.
Publication year - 2010
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.2010.05.042
Subject(s) - dementia , cognitive impairment , neuroimaging , hippocampus , alzheimer's disease , nuclear medicine , magnetic resonance imaging , alzheimer's disease neuroimaging initiative , medicine , psychology , neuroscience , cardiology , disease , radiology
Results: Consideration of both baseline and follow-up data provides increased classification accuracies, determined using linear discriminant analysis, compared with those obtained using the baseline data alone. Accuracy increased from 72% to 78% between AD patients and HC, 65% to 68% between aMCI patients and HC, and 58% to 61% between AD and aMCI patients. Conclusions: This work-in-progress follows from the successful application of this regional analysis technique to baseline FDG-PET data from the ADNI, in which the features used for classification include not only the hippocampus, but all those extracted from a subject-specific segmentation consisting of 83 anatomical regions. These early results suggest that the group discrimination achieved using baseline data alone may be further improved by the inclusion of follow-up FDG-PET data.