Premium
P1‐366: Advances in individually‐based imaging biomarkers using multi‐parametric pattern classification
Author(s) -
Davatzikos Christos
Publication year - 2011
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.2011.05.646
Subject(s) - categorical variable , artificial intelligence , cluster analysis , parametric statistics , cognition , disease , pattern recognition (psychology) , computer science , machine learning , medicine , psychology , neuroscience , pathology , mathematics , statistics
(LO) fNIRS. 14 patients and 49 controls completed a follow-up measurement after one year. Results: During the letter version of the VFT and the LO AD-patients displayed a reduced activation pattern in the prefrontal (VFT) and the parietal cortex (LO) as compared to healthy controls. Increases in O2Hb in parietal regions correlated with prefrontal activity in AD-patients but remained independent in healthy controls. Prefrontal activation decreased significantly over the one year interval in AD-patients. Healthy controls showed no changes at follow up. Activation decreases in the prefrontal cortex were correlated with the decrease of the MMSE score, with greater decrease indicating lower MMSE score at the followup. Conclusions: fNIRS seems to be a well suited instrument to measure disease related changes in the cortex of AD patients and their progression over time. Considering the fact that we did not observe any changes in healthy controls without any cognitive decline fNIRS could be a promising device to be used in prospective studies of AD and other dementias.