Weighted Brain Network Analysis on Different Stages of Clinical Cognitive Decline

Area of Science:

• Neuroscience
• Cognitive Science
• Biophysics

Background:

• Cognitive dysfunction encompasses a spectrum from subjective cognitive impairment (SCI) to mild cognitive impairment (MCI) and Alzheimer's disease (AD).
• Understanding brain network alterations across these stages is crucial for early diagnosis and intervention.
• Electroencephalography (EEG) provides a valuable tool for analyzing brain activity and functional connectivity.

Purpose of the Study:

• To analyze brain network topology across different clinical severity stages of cognitive dysfunction (SCI, MCI, AD) using EEG.
• To investigate the spatiotemporal dynamics of functional connectivity and network properties.
• To identify early alterations in brain networks associated with cognitive decline.

Main Methods:

• Utilized EEG data from patients diagnosed with SCI, MCI, and AD.
• Developed a novel framework employing spatiotemporal entropy to estimate functional connectivity.
• Applied graph theory analysis, including clustering coefficient, path length, and modularity, across different frequency bands (delta, alpha).

Main Results:

• Functional connectivity and graph analysis findings are dependent on the EEG frequency band.
• Significant network alterations are detectable at the MCI stage, indicating a transition.
• SCI networks demonstrated increased resilience, with distinct topological properties in delta and alpha bands compared to MCI and AD.
• Modularity analysis revealed unique anterior-posterior network organization in SCI.
• MCI exhibited high-strength intrinsic connectivity, suggesting early compensatory mechanisms.

Conclusions:

• Brain network alterations in cognitive dysfunction are frequency-band specific and emerge early, starting at the MCI stage.
• Subjective cognitive impairment (SCI) networks show greater resilience to neuronal damage compared to MCI and AD.
• Mild cognitive impairment (MCI) represents a transitional phase between SCI and AD, characterized by compensatory network activity.