Improved classification of Alzheimer's disease and mild cognitive impairment through dynamic functional network analysis
About
Brain networks from functional MRI have advanced our understanding of cortical activity and its disruption in neurodegenerative disorders. Recent work has increasingly focused on dynamic (time-varying) brain networks that capture both spatial and temporal patterns of regional co-activity, yet this approach remains underexplored across the Alzheimer's disease (AD). We analysed age- and sex-matched static and dynamic functional brain networks derived from resting-state fMRI data in 315 individuals with AD, mild cognitive impairment (MCI), and cognitively normal healthy controls (HC) from the ADNI-3 cohort. Functional networks were constructed using the Juelich brain atlas, with static connectivity estimated from full time series and dynamic connectivity derived using a sliding-window approach. Group differences were assessed at both link and node levels using non-parametric statistics and bootstrap resampling. While HC and MCI exhibited similar static and dynamic patterns at the node level, clearer differences emerged in AD. Stable (stationary) differences in functional connectivity were identified between white matter regions and parietal and somatosensory cortices, whereas temporally varying differences were consistently observed in connections involving the amygdala and hippocampal formation. Node centrality analysis further suggested that white matter connectivity differences are predominantly local in nature. These findings highlight both shared and distinct functional connectivity patterns across static and dynamic networks, underscoring the importance of incorporating temporal dynamics into brain network analyses of the Alzheimer's spectrum. Additionally, a Random Forest model trained on regional BOLD time series informed by static and dynamic metrics achieved robust classification of MCI, AD, and HC groups, demonstrating the diagnostic potential of time-varying connectivity.