Spatiotemporal Network Dynamics Reveal Alzheimer's Disease Progression.

Area of Science:

• Neuroscience
• Medical Imaging
• Biomarker Discovery

Background:

• Alzheimer's disease (AD) involves progressive brain network disruptions preceding cognitive decline.
• Conventional functional connectivity analyses often miss critical early network coordination changes.
• Identifying preclinical AD biomarkers is crucial for timely intervention and treatment.

Purpose of the Study:

• To characterize spatiotemporal network alterations in Alzheimer's disease (AD) using novel analytical techniques.
• To identify early network changes indicative of preclinical AD progression.
• To establish a dynamic, network-level biomarker framework for AD.

Main Methods:

• Utilized resting-state fMRI data from the Alzheimer's Disease Neuroimaging Initiative (ADNI) cohorts.
• Applied quasi periodic patterns (QPPs) to derive spatiotemporal templates and network integrity.
• Employed complex principal component analysis (cPCA) to analyze amplitude and phase relationships in brain activity.

Main Results:

• Identified a structured trajectory of network degradation, with early impact on limbic, subcortical, and higher cognition networks.
• Observed significant alterations in transitioning cohorts before formal diagnostic conversion, indicating preclinical information.
• QPP metrics highlighted limbic/subcortical degradation, while cPCA revealed changes in higher-order, visual, and cerebellar networks.

Conclusions:

• Combining QPP and cPCA provides complementary insights into AD-related network pathology.
• Spatiotemporal signatures detected by these methods offer potential for early AD detection.
• The developed framework can aid in characterizing disease trajectories and monitoring treatment efficacy.