Premium
IC‐P‐140: Associations between years of education, Aβ deposition, and metabolism in cognitively normal older adults, mild cognitive impairment, and Alzheimer's disease: Evidence for reserve
Author(s) -
Arenaza-Urquijo Eider M.,
Wirth Miranka,
Gonneaud Julie,
Bejanin Alexandre,
Mutlu Justine,
Mézenge Florence,
Landeau Brigitte,
Sayette Vincent,
Eustache Francis,
Chételat Gaël
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.2015.06.162
Subject(s) - clinical dementia rating , cognition , pittsburgh compound b , cognitive reserve , neuroimaging , psychology , dementia , alzheimer's disease , cognitive decline , alzheimer's disease neuroimaging initiative , disease , cognitive impairment , audiology , medicine , neuroscience
ADNI dataset. 17 features served as inputs to the classifier: three MRI intensities (T1-w, T2*, and FLAIR), spatial probability map, intensity distribution probabilities for healthy and WMH tissues (PWM,PWMH) as well as the ratio PWM/PWMH for each modality, and the average intensity of healthy tissue at each voxel for eachmodality. WMHs were segmented using a linear regression classifier with thresholding yielding low variance, high accuracy and low computation time. Performance was evaluated with 10-fold crossvalidation, using Dice Kappa similarity, sensitivity, and specificity to compare automated and manual labels. Results: The mean Dice Kappa, sensitivity, and specificity between automated and manual labels were 0.46, 0.55, and 0.999, respectively. For further validation, total WMH loads were correlated with Fazekas scores of 30 subjects (r1⁄40.80, p1⁄46.44e-7 for periventricular and r1⁄40.55, p1⁄42.8e-3 for deep WMHs) as well as age for the total population (r1⁄40.43, p1⁄41.52e-6). Figures 1.a., 1.b. and 1.c. show the FLAIR images, and automated and manual labels for a subject. Figures 1.d. and 1.e. show Kappa and sensitivity/specificity versus threshold. Conclusions:Automated labels showed reasonable agreement with manual labels and had significant correlations with Fazekas ratings. Our results therefore suggest that the proposed automated tool can provide fast, robust, and accurate labels for WMHs and holds good potential for clinical studies. (Funding: CIHR MOP-111169, Pfizer Canada, FRQ-S, and Douglas Hospital Research Centre)