Journal Article

Group and Individual fMRI Analysis of the Main Resting State Networks in Healthy Subjects

Sharova, E V, Boldyreva, G N, Pronin, I N, Deza-Araujo, Y I, Kulikov, M A, Knyazev, G G, Verkhlyutov, V M, Gavron A A, Smirnov, A S, Fadeeva, L M, Abdulaev, A A, Zhavoronkova, L A, Chelyapina, M V

2020

Overview

Analysis of fMRI in the resting state (RS) is a suitable methodological approach to studying basal levels of functional brain activity in humans in health and disease. The inadequate development of this direction in Russia is partly due to the small number of Russian publications describing approaches to data processing. This study uses an algorithm for analysis of fMRI signals in the RS based on independent components analysis (ICA) run in the FSL environment and used for studies of typical functional resting state networks (RSN) in health. Averaging observation data by group, which is applicable for studies of healthy people, is often not appropriate for studies of different forms of cerebral pathology, which are characterized by significantly greater levels of variation in hemodynamics. Thus, studies of 17 healthy subjects included comparative evaluation of the topography and a number of quantitative measures of typical RSN identified by group and individual analysis of fMRI signals. These networks were comparable with RSN described in the literature as main and were also reproducible in group and individual analysis, which confirms the suitability, reliability, and effectiveness of using this algorithm. Individual analysis of RSN identified variability linked with a number of psychophysiological characteristics of healthy subjects (sex, motor asymmetry profile, EEG pattern), partly explaining the different levels of compliance with the patterns of the group networks. Results obtained from individual fMRI and EEG comparisons showed the potential of analysis of the topography of the sources of individual rhythms as EEG markers for RSN. The lowest levels of variability of fMRI characteristics of resting networks in health (such as maximum network activation intensity, mean frequency of the active zone of the spectrum, frequency of dominant peak) may have diagnostic value for studies of RSN in pathology.

• Publisher: Springer Nature B.V
• Subject: Algorithms / Brain mapping / EEG / Functional magnetic resonance imaging / Hemodynamics / Topography