Phase shift invariant imaging of coherent sources (PSIICOS) from MEG data

Area of Science:

• Neuroscience
• Computational Neuroscience
• Signal Processing

Background:

• Neuronal communication relies on synchronized activity across brain regions.
• Conventional EEG/MEG connectivity metrics suffer from spatial leakage, obscuring true interactions.
• Existing methods fail to accurately detect zero- or near-zero phase lag synchrony.

Purpose of the Study:

• To develop a novel method for mitigating spatial leakage in electroencephalography (EEG) and magnetoencephalography (MEG) data.
• To enable accurate detection of neuronal coupling, including zero- and near-zero phase lag interactions.
• To introduce the Phase Shift Invariant Imaging of Coherent Sources (PSIICOS) method for improved brain connectivity analysis.

Main Methods:

• A projection operation on sensor-space cross-spectrum to suppress spatial leakage.
• Network estimation framed as a source estimation problem in a product space of source topographies.
• Validation through analytical derivations, realistic simulations, and application to real MEG data.

Main Results:

• The proposed method, PSIICOS, effectively mitigates spatial leakage.
• PSIICOS demonstrates superior detection characteristics compared to existing metrics in simulations.
• The method successfully identified true zero-phase coupling in a real MEG dataset during a mental rotation task.

Conclusions:

• PSIICOS offers a robust solution for estimating brain connectivity from EEG/MEG data.
• This method overcomes limitations of previous approaches, particularly in detecting zero-phase interactions.
• PSIICOS advances non-invasive electrophysiological recordings for understanding neuronal communication and brain networks.