Partial directed coherence-based information flow in Parkinson's disease patients performing a visually-guided motor...

Area of Science:

• Neuroscience
• Biomedical Engineering
• Computational Neuroscience

Background:

• Parkinson's disease (PD) is a neurodegenerative disorder affecting motor control.
• Understanding information flow in the brain is crucial for diagnosing and treating PD.
• Existing methods may not capture dynamic connectivity changes effectively.

Purpose of the Study:

• To propose and validate a novel method for analyzing brain connectivity in PD patients.
• To investigate alterations in information flow during a motor task in PD.
• To assess the impact of medication on brain connectivity in PD.

Main Methods:

• Utilized a sparse multivariate autoregressive (mAR) model to estimate partial directed coherence (PDC).
• Applied the sparse-PDC method to electroencephalography (EEG) data from PD patients and healthy controls.
• Analyzed information flow patterns during a visually-guided motor task.

Main Results:

• Demonstrated significantly altered patterns of brain connectivity in PD patients compared to controls.
• Identified altered net information outflow from the left sensorimotor region in PD patients.
• Observed partial restoration of disrupted connectivity patterns after levodopa medication.
• Showed sparse-PDC is more sensitive to dynamic connectivity changes than traditional spectral analysis.

Conclusions:

• The proposed sparse-PDC method is effective for investigating altered brain connectivity in PD.
• Information flow disruptions in PD are linked to specific brain regions and are modulated by medication.
• This technique offers a sensitive approach to capture dynamic changes in neural communication.