Complexity of brain activity and connectivity in functional neuroimaging

Area of Science:

• Neuroscience
• Computational Neuroscience
• Signal Processing

Background:

• Understanding human brain dynamics and connectivity is crucial across various conditions.
• Novel measures are needed for multichannel activity and dynamic functional brain connectivity (DFBC).

Purpose of the Study:

• To introduce a novel complexity index (CI) and flexibility index (FI) for quantifying brain activity and connectivity dynamics.
• To analyze DFBC using established and novel metrics across multiple neuroimaging modalities.

Main Methods:

• Defined a complexity index (CI) from symbolic dynamics to quantify brain activity patterns.
• Analyzed DFBC using sliding window approach with imaginary part of phase locking value (iPLV) for EEG/ECoG/MEG and wavelet coherence (WC) for fMRI.
• Estimated intra- and cross-frequency couplings (CFC) and defined dominant intrinsic coupling modes (DICM) per region-of-interest (ROI).
• Introduced a flexibility index (FI) to quantify transitions in DICM between time windows.

Main Results:

• Successfully discriminated healthy controls from schizophrenic patients using FI and DICM dynamic reconfiguration.
• Demonstrated that anesthesia globally decreased complexity across frequency bands (except δ) and altered DICM dynamic reconfiguration.
• Showcased high reliability of CI and DICM in MEG/fMRI resting-state recordings at two spatial scales.

Conclusions:

• The novel CI and FI provide robust measures for brain complexity and dynamic connectivity.
• These indices offer valuable insights into neurophysiological changes associated with clinical conditions and interventions.
• The methodology is broadly applicable across various neuroimaging modalities and research questions.