Premium
O2–03–01: Individual patient diagnosis of Alzheimer's and frontotemporal dementias via high‐dimensional pattern classification of MRI
Author(s) -
Davatzikos Christos,
Resnick Susan M.,
Wu Xiaoying,
Parmpi Paraskevi,
Clark Christopher M.
Publication year - 2007
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.2007.04.026
Subject(s) - frontotemporal dementia , atrophy , magnetic resonance imaging , voxel , dementia , alzheimer's disease , neuroimaging , psychology , medicine , diagnostic accuracy , neuroscience , pathology , disease , radiology
whole brain. The regional hippocampal connectivity map (rHCM) was obtained by group t-test between the MCI and CN groups. The regional hippocampal connectivity index (rHCI) was then obtained by averaging the absolute value of cross-correlation coefficients within the rHCM. Results: The rHCM included the cunes, precunes, anterior and posterior cingulate gyrus, bilateral medial frontal gyrus, superior temporal gyrus, parahippocampal gyrus, fusiform gyrus, and right inferior parietal lobe. The rHCI for the CN (0.386 0.012) was significantly lower than that of the MCI (0.359 0.007) using t-test (p 10) and the rHCI for the rMCI (0.379 0.009) was in between, as shown in Figure 1. Figure 2 shows that the rHCI was significantly correlated with the Rey Auditory Verbal Learning (RAVLT) (R 0.38, p 0.0001) neuropsychological test scores. Conclusions: This study found that the subjects with AD risk show significant loss of functional connectivity in a set of regions from MTL. The rHCI separates MCI from CN with a high accuracy and the rHCI is highly correlated to neuropsychological test scores. Our results suggest that the rHCI could potentially serve as a noninvasive quantitative marker for the preclinical stage of early AD. MONDAY, JUNE 11, 2007 ORAL O2-03 EARLY DETECTION AND DIAGNOSIS 4