Effective differentiation of mild cognitive impairment by functional brain graph analysis and computerized testing

Area of Science:

• Neuroscience
• Gerontology
• Medical Imaging

Background:

• African American elders face a doubled risk of mild cognitive impairment (MCI) and Alzheimer's disease compared to white elders.
• Current advanced diagnostic methods like imaging and cerebrospinal fluid analysis present significant cost and accessibility barriers for this population.

Purpose of the Study:

• To investigate the efficacy of resting-state electroencephalography (EEG) functional connectivity and graph theory measures, combined with cognitive testing, for detecting MCI in community-dwelling African American elders.
• To establish a cost-effective and accessible method for early MCI detection in a high-risk demographic.

Main Methods:

• Utilized resting-state electroencephalography (EEG) to derive functional connectivity and graph theoretical measures.
• Administered computerized cognitive tests to assess cognitive function.
• Employed a machine learning approach to differentiate between individuals with MCI and healthy controls using combined EEG and cognitive data.

Main Results:

• Found significant reductions in functional connectivity and graph topology integration in individuals with MCI.
• Achieved a prediction accuracy of 86.5% for MCI when combining cognitive, functional connectivity, and topological features.
• Demonstrated superior predictive power compared to single-domain approaches (best single approach accuracy: 77.5%).

Conclusions:

• A combined approach utilizing EEG-derived measures and cognitive testing offers a powerful and accurate method for predicting MCI in African American elders.
• Resting-state EEG and computerized cognitive testing are acceptable and promising tools for early MCI detection in community-dwelling older adults.
• This methodology addresses the need for accessible and cost-effective early detection of cognitive decline in at-risk populations.