A novel method on functional connectivity biomarkers: the potential of spatiotemporal graph convolutional network in identifying imaging biomarkers for Alzheimer’s disease

The abnormal deposition of pathological proteins in Alzheimer’s disease (AD) can disrupt neural pathways in the brain. Functional magnetic resonance imaging (fMRI) is a valuable tool for detecting changes in brain connectivity caused by AD, offering significant support for early diagnosis. However, traditional numerical distribution methods often struggle to capture subtle lesions in fMRI images, limiting their diagnostic utility. The study discussed in this commentary introduces a spatiotemporal graph convolutional network (ST-GCN) framework that integrates both spatial and temporal dependencies in brain functional networks, providing a more robust approach to fMRI data analysis. This advancement significantly improves the identification of functional connectivity (FC) biomarkers for AD. This commentary critically evaluates the strengths and limitations of the proposed approach and outlines potential future research directions, particularly in the integration of multimodal neuroimaging data and further validation of the model across diverse clinical populations.