TY - JOUR
T1 - Alzheimer's Imaging Consortium
AU - Meysami, Somayeh
AU - Raji, Cyrus A.
AU - Lee, Soojin
AU - Garg, Saurabh
AU - Akbari, Nasrin
AU - Pompa, Rodrigo Solis
AU - Gouda, Ahmed
AU - Nguyen, Thanh Duc
AU - Basar, Saqib
AU - Chodakiewitz, Yosef Gavriel
AU - Merrill, David A.
AU - Patel, Amar
AU - Durand, Daniel J.
AU - Hashemi, Sam
N1 - Publisher Copyright:
© 2025 The Alzheimer's Association. Alzheimer's & Dementia published by Wiley Periodicals LLC on behalf of Alzheimer's Association.
PY - 2025/12/1
Y1 - 2025/12/1
N2 - BACKGROUND: Brain age - an estimate of chronological age- derived from structural brain MR neuroimaging may reveal underlying factors driving brain aging. White matter hyperintensities (WMH) are areas of abnormally high signal on FLAIR that frequently reflect chronic small vessel ischemic changes and potentially increased aging. METHOD: Overall, 1,164 healthy participants from four sites (mean chronological age 55.17 ± 12.37 years, 52% women; 48% men; 39% non-white) were scanned on 1.5T MR machines with a whole-body protocol. For each participant, a 2D multi-slice FLAIR image was obtained. A 2D convolutional neural network, trained on data from 120 individuals across three public datasets (MICCAI 2017, ISLES2015, and ISLES2022), was employed to automatically segment WMH from the FLAIR scans. Deep learning with FastSurfer on MPRAGE trained on 134 participants aged 27-66 and segmented 96 brain regions. Brain age was computed using a regression-based 3D Simple Fully Convolutional Network trained on in-house T1-weighted MRI scans collected from 5,500 healthy individuals (, aged 18 to 89 years). Brain age gap (BAG) was computed by subtracting chronological age from brain age. Partial correlation and regression models evaluated the relationship between WMH normalized to total brain volume (gray matter and white matter), brain age, brain volumes controlling for age, sex, and total intracranial volume. Chronological age was not adjusted for in the brain age models to avoid collinearity. RESULT: Mean brain age was similar to chronological age (mean brain age = 56.04 ± 12.65, mean BAG = 0.69). The median of WMH were 1.4 mL (0.75-2.51 mL). Increased WMH were related to lower brain volumes in the i) hippocampus (rp= -0.13, p = 1.174e-05) ii) cerebral white matter (rp= -0.08, p = .004) iii) thalamus (rp= -0.16, p = 1.634e-07). Additionally, increased WMH was related to larger cerebral ventricle size (rp= 0.28, p = 6.604e-21). The regression model showed that increased WMH was related to increased brain age (t= 5.92, rp= .42, p <.001) and increased brain age gap (t= 4.07, rp= .12, p <.001). CONCLUSION: Increased WMH are related to brain atrophy - in both Alzheimer's and non-Alzheimer's affected regions - and are also related to increased brain age and accelerated brain aging.
AB - BACKGROUND: Brain age - an estimate of chronological age- derived from structural brain MR neuroimaging may reveal underlying factors driving brain aging. White matter hyperintensities (WMH) are areas of abnormally high signal on FLAIR that frequently reflect chronic small vessel ischemic changes and potentially increased aging. METHOD: Overall, 1,164 healthy participants from four sites (mean chronological age 55.17 ± 12.37 years, 52% women; 48% men; 39% non-white) were scanned on 1.5T MR machines with a whole-body protocol. For each participant, a 2D multi-slice FLAIR image was obtained. A 2D convolutional neural network, trained on data from 120 individuals across three public datasets (MICCAI 2017, ISLES2015, and ISLES2022), was employed to automatically segment WMH from the FLAIR scans. Deep learning with FastSurfer on MPRAGE trained on 134 participants aged 27-66 and segmented 96 brain regions. Brain age was computed using a regression-based 3D Simple Fully Convolutional Network trained on in-house T1-weighted MRI scans collected from 5,500 healthy individuals (, aged 18 to 89 years). Brain age gap (BAG) was computed by subtracting chronological age from brain age. Partial correlation and regression models evaluated the relationship between WMH normalized to total brain volume (gray matter and white matter), brain age, brain volumes controlling for age, sex, and total intracranial volume. Chronological age was not adjusted for in the brain age models to avoid collinearity. RESULT: Mean brain age was similar to chronological age (mean brain age = 56.04 ± 12.65, mean BAG = 0.69). The median of WMH were 1.4 mL (0.75-2.51 mL). Increased WMH were related to lower brain volumes in the i) hippocampus (rp= -0.13, p = 1.174e-05) ii) cerebral white matter (rp= -0.08, p = .004) iii) thalamus (rp= -0.16, p = 1.634e-07). Additionally, increased WMH was related to larger cerebral ventricle size (rp= 0.28, p = 6.604e-21). The regression model showed that increased WMH was related to increased brain age (t= 5.92, rp= .42, p <.001) and increased brain age gap (t= 4.07, rp= .12, p <.001). CONCLUSION: Increased WMH are related to brain atrophy - in both Alzheimer's and non-Alzheimer's affected regions - and are also related to increased brain age and accelerated brain aging.
UR - https://www.scopus.com/pages/publications/105025735781
U2 - 10.1002/alz70862_110308
DO - 10.1002/alz70862_110308
M3 - Article
C2 - 41433484
AN - SCOPUS:105025735781
SN - 1552-5260
VL - 21
SP - e110308
JO - Alzheimer's & dementia : the journal of the Alzheimer's Association
JF - Alzheimer's & dementia : the journal of the Alzheimer's Association
ER -