Reliability of fully automated versus visually controlled pre- and post-processing of resting-state EEG

Area of Science:

• Neuroscience
• Signal Processing

Background:

• Resting-state electroencephalography (EEG) analysis is crucial for understanding brain function.
• Traditional visual analysis (VA) of EEG data is time-consuming and subjective.
• Developing automated methods for EEG processing can improve efficiency and standardization.

Purpose of the Study:

• To compare the reliability of a novel automated analysis (AA) toolbox for resting-state EEG with traditional visual analysis (VA).
• To evaluate the consistency of AA and VA over time using intraclass correlation coefficients (ICC).

Main Methods:

• 34 healthy volunteers underwent three resting-state EEG recordings at one-year intervals.
• EEG data were processed using both automated analysis (AA) and visual analysis (VA) methods.
• Frequency analysis results were compared using Pearson correlation coefficients, and reliability was assessed with ICC.

Main Results:

• The mean correlation coefficient between AA and VA was high (0.94±0.07), indicating strong agreement.
• Mean ICC for AA was 0.83±0.05, and for VA was 0.84±0.07, demonstrating comparable reliability over time.
• Both methods yielded very similar results for spectral EEG analysis.

Conclusions:

• Automated analysis (AA) and visual analysis (VA) provide comparable results for spectral EEG analysis.
• AA offers significant advantages, including reduced time, complete standardization, and independence from rater bias.
• Automated EEG processing streamlines quantitative EEG analysis workflows.