PhD Scientific Days 2020

Neurosciences II. Lectures

Bivariate Focus-Based Multifractal Formalism: A Novel Method for Estimating the Multifractal Dynamics of the Resting State Functional Connectivity

Előadó neve

Dr. Stylianou, Orestis

Előadó munkahelye

Department of Physiology, Semmelweis University

Előadó telefonszáma

+36702731655

Előadó e-mail címe

orestisstylianou@rocketmail.com

Az előadás címe

Bivariate Focus-Based Multifractal Formalism: A Novel Method for Estimating the Multifractal Dynamics of the Resting State Functional Connectivity

Szerző(k) neve és munkahelye

Orestis Stylianou MD, Department of Physiology, Semmelweis University, Budapest
Peter Mukli MD PhD, Department of Physiology, Semmelweis University, Budapest
Frigyes Sammuel Racz MD PhD, Department of Physiology, Semmelweis University, Budapest

Szekció

Neurosciences II. Lectures

Language of the presentation

English

Section, first choice

Neurosciences

Section, second choice

Theoretical and Translational Medicine

Összefoglaló szövege

Introduction: Lately, examining resting-state (rs) brain network dynamics – via functional connectivity (FC) of interacting regions – has been a growing field of neuroscience. Such investigations can be also carried out by electroencephalography (EEG) that captures the underlying rapid neuronal dynamics from functionally coupled cortical areas. As a result of previous developments, the interplay between resting-state FC and scale-free brain dynamics could be characterized with the aid of bivariate multifractal (MF) analytical tools. However, a robust characterization of the scale-free nature of the coupled EEG-fluctuations is still lacking. This scarcity gave birth to the bivariate focus-based multifractal formalism (BFMF).
Aims: The purpose of this work was to demonstrate the valuable features of this novel method by investigating the presence of MF dynamics as well as their spatial organization in the rs EEG.
Methods: EEG of 12 subjects was recorded during 5 minutes of eyes closed in rs using a 62-channel BrainAmp. Based on the covariance between each pair of EEG signals, calculated for a set of time scales, BFMF estimated the scale-free exponent function H(q) for various order parameters (-15≤q≤15). From H(q), a measure of long-term memory (H(2)) and a non-linearity parameter (ΔH15) were obtained for every time series pair. Diverse tests were realized for the verification of the true multifractality in each pair. Having the EEG channels grouped into 6 resting state networks (RSNs), the MF measures captured connectivity within and between RSNs. Kendall’s W examined the subject concordance while individual t-tests examined the variability of within and between RSNs connections. Finally, the MF functional networks were compared with networks constructed using Pearson correlation and mutual information (MI) as FC estimators.
Results: Most of coupled EEG-dynamics showed true MF character. Regional variability of connections as well as subject agreement was found to be significant. When compared to the Pearson and MI constructed networks the MF networks differed in their architecture.
Conclusion: This study provides insight into the spatial organization of brain networks by utilizing BFMF, a method capable of capturing the scale-free coupling between EEG time series. BFMF could provide the clinical neuropsychiatric studies with new, much needed, biomarkers.

Additional Information

Peter Mukli
E-mail: mukli.peter@med.semmelweis-univ.hu

Bemutatás módja

Szóbeli

Kind

Szabad

Status

elfogadva

Accepted presentation method

szóbeli

Előadás fájl jóváhagyás

nem rendelkezett róla

Előadó

4605

Start

17:45

End

18:00

Authors (legacy)

Orestis Stylianou MD, Department of Physiology, Semmelweis University, Budapest
Peter Mukli MD PhD, Department of Physiology, Semmelweis University, Budapest
Frigyes Sammuel Racz MD PhD, Department of Physiology, Semmelweis University, Budapest