Complexity measures in magnetoencephalography: measuring "disorder" in schizophrenia

Area of Science:

• Neuroscience
• Computational Neuroscience
• Biophysics

Background:

• Magnetoencephalography (MEG) is a non-invasive technique to measure brain activity.
• Understanding the spatio-temporal dynamics of brain 'disorder' is crucial for neuroscience.
• Existing methods like time-frequency decomposition offer insights into neural oscillations.

Purpose of the Study:

• To develop and validate a novel methodology for measuring spatio-temporal brain dynamics using signal entropy.
• To explore the relationship between signal entropy and established neural oscillation metrics.
• To demonstrate the clinical utility of the entropy-based method in characterizing neurophysiological abnormalities in schizophrenia.

Main Methods:

• Application of a signal entropy-based methodology to magnetoencephalography (MEG) data.
• Analysis of spatio-temporal dynamics and independence of entropy time-courses across brain regions.
• Comparison of entropy measures with time-frequency decomposition techniques.
• Assessment of task-induced entropy changes in healthy controls and schizophrenia patients.

Main Results:

• Spatially distinct brain regions exhibit temporally independent entropy time-courses.
• Cognitive tasks modulate entropy, with increased local neural processing shown as transient entropy increases.
• A complex, complementary relationship exists between signal entropy and oscillatory amplitude.
• Schizophrenia patients show significantly increased task-induced entropy changes in specific brain networks, including the salience network.

Conclusions:

• The developed signal entropy methodology effectively measures brain signal dynamics.
• Entropy and oscillatory amplitude are complementary metrics for neural signal analysis.
• The findings support the hypothesis of salience network abnormalities in schizophrenia and suggest clinical utility for the entropy-based method.