INTRODUCTIONThis study aimed to explore the potential of whole brain white matter patterns as novel neuroimaging biomarkers for assessing cognitive impairment and disability in older adults.METHODSWe conducted an in‐depth analysis of magnetic resonance imaging (MRI) and amyloid positron emission tomography (PET) scans in 454 participants, focusing on white matter patterns and white matter inter‐subject variability (WM‐ISV).RESULTSThe white matter pattern ensemble model, combining MRI and amyloid PET, demonstrated a significantly higher classification performance for cognitive impairment and disability. Participants with Alzheimer's disease (AD) exhibited higher WM‐ISV than participants with subjective cognitive decline, mild cognitive impairment, and vascular dementia. Furthermore, WM‐ISV correlated significantly with blood‐based biomarkers (such as glial fibrillary acidic protein and phosphorylated tau‐217 [p‐tau217]), and cognitive function and disability scores.DISCUSSIONOur results suggest that white matter pattern analysis has significant potential as an adjunct neuroimaging biomarker for clinical decision‐making and determining cognitive impairment and disability.Highlights
The ensemble model combined both magnetic resonance imaging (MRI) and amyloid positron emission tomography (PET) and demonstrated a significantly higher classification performance for cognitive impairment and disability.
Alzheimer's disease (AD) revealed a notably higher heterogeneity compared to that in subjective cognitive decline, mild cognitive impairment, or vascular dementia.
White matter inter‐subject variability (WM‐ISV) was significantly correlated with blood‐based biomarkers (glial fibrillary acidic protein and phosphorylated tau‐217 [p‐tau217]) and with the polygenic risk score for AD.
White matter pattern analysis has significant potential as an adjunct neuroimaging biomarker for clinical decision‐making processes and determining cognitive impairment and disability.