Predicting Dementia With Prefrontal Electroencephalography and Event-Related Potential

Area of Science:

• Neuroscience
• Gerontology
• Medical Diagnostics

Background:

• Cognitive decline and dementia pose significant public health challenges.
• Early and accurate screening methods are crucial for timely intervention and management.
• Current screening tools like the Mini-Mental Status Examination (MMSE) have limitations in sensitivity and specificity.

Purpose of the Study:

• To evaluate the efficacy of prefrontal electroencephalography (EEG) markers for dementia screening.
• To compare the predictive performance of EEG/event-related potential (ERP) variables against the MMSE.
• To investigate the combined utility of EEG/ERP data and MMSE for improving dementia prediction.

Main Methods:

• Utilized resting-state EEG, sensory ERP, and selective-attention ERP in 122 elderly participants (35 with dementia, 87 controls).
• Assessed correlations between EEG/ERP variables and MMSE scores using Pearson's correlation and linear regression.
• Employed logistic regression and cross-validation with classification algorithms to develop and validate predictive models for dementia.

Main Results:

• Prefrontal EEG/ERP variables demonstrated significant correlations with cognitive decline (MMSE scores).
• Selective-attention ERP and resting-state EEG variables showed superior predictive performance for dementia compared to MMSE alone.
• Combining EEG/ERP markers with MMSE scores significantly improved predictive accuracy, outperforming demographic factors.

Conclusions:

• Prefrontal EEG markers are effective for dementia screening, individually and in conjunction with MMSE.
• EEG-based methods offer a promising alternative or adjunct to traditional cognitive assessments for dementia detection.
• Further research into EEG/ERP applications can advance early diagnosis and management of dementia.