A morphological way to remove baseline and spike separation in EEG

Area of Science:

• Neuroscience
• Signal Processing
• Biomedical Engineering

Background:

• Independent Component Analysis (ICA) is widely used for electroencephalogram (EEG) signal separation.
• ICA methods can suffer from inconsistent decomposition quality and ordering of source signals.
• Accurate source separation is crucial for understanding brain activity and diagnosing neurological conditions.

Purpose of the Study:

• To address the limitations of ICA in EEG signal decomposition.
• To introduce and evaluate Morphological Component Analysis (MCA) as an alternative for EEG source separation.
• To compare the performance of MCA and ICA in segregating specific EEG signal components.

Main Methods:

• Utilized Morphological Component Analysis (MCA) with an explicit dictionary of independent redundant bases.
• Applied MCA to separate source signals from electroencephalogram (EEG) data.
• Qualitatively compared MCA with Independent Component Analysis (ICA) using correlation analysis.

Main Results:

• MCA provides a more consistent approach to decomposing EEG signals compared to ICA.
• MCA demonstrated effectiveness in segregating baseline EEG activity.
• MCA showed superior capability in isolating transient spike events in EEG data.

Conclusions:

• Morphological Component Analysis (MCA) presents a viable and potentially more robust alternative to ICA for EEG source signal separation.
• MCA's dictionary-based approach offers improved consistency and quality in signal decomposition.
• Further research can explore quantitative comparisons and applications of MCA in clinical EEG analysis.