12/18/2023 0 Comments Negative correlationAlzheimer’s disease (AD) has been consistently characterized by decreased FC in alpha frequency band ( Koenig et al., 2005 Wang et al., 2014 Babiloni et al., 2016). However, few studies have also found a decrease in alpha FC ( Shim et al., 2018 Zhang et al., 2018). Major depressive disorder (MDD) is mostly characterized by increased FC ( Fingelkurts et al., 2007 Leuchter et al., 2012 Olbrich et al., 2014 Li et al., 2017) and more random network structure ( Li et al., 2017 Zhang et al., 2018 Sun et al., 2019) in theta and alpha frequency bands. Therefore, we will focus on frequency bands, where the most frequent and consistent results were reported. Previous studies have found results in all frequency bands, but often inconsistencies between studies occur. Since then, studying small-world properties of functional brain networks has been widely used.Ĭhanges in EEG resting state FC and small-world structure are often used for statistical analysis between two populations, generally with the aim to compare patient and control groups. A measure of small-worldness (SW) has been proposed to assess small-world properties of a network ( Humphries and Gurney, 2008). In that case, functional integration and functional segregation are simultaneously high. A network is compared to random networks and in order to have small-world properties, the network should be more clustered than a random network, but have similar characteristic path length ( Watts and Strogatz, 1998 Albert and Barabási, 2002 Rubinov and Sporns, 2010 Bassett and Bullmore, 2017). Small-world organization is one of the most frequently analyzed topological properties of functional neural networks. Graphs are constructed by removing edges with lowest values. Real-life neural networks are represented graphically, using electroencephalographic (EEG) channels as nodes and FC as edges between nodes. Complex network analysis is based on classical graph theoretical analysis, but focuses on analyzing complex real-life networks ( Rubinov and Sporns, 2010). Significant work has been done to show that neural network architecture can be adaptively reconfigured between different states of the subjects ( Bassett et al., 2006 Liu et al., 2015a Lin et al., 2020) and associate network topology to physiologic states ( Bashan et al., 2012 Bartsch and Ivanov, 2014 Ivanov and Bartsch, 2014 Bartsch et al., 2015 Liu et al., 2015b).įunctional connectivity and complex network analysis have been the most widely used types of brain network analysis by providing the tools to analyze the brain as a network of interacting regions, while maintaining computational simplicity. Functional connectivity is crucial also in brain physiology ( Lynn and Bassett, 2019). Based on the results of the current study, we suggest that decreased alpha small-world organization is compensated with increased connectivity of alpha oscillations in a healthy brain.įunctional connectivity (FC) is highly important in physiology at various levels: from molecules to organs and physiological networks are not only of wide scientific interest, but also have high impact in medicine ( Ivanov et al., 2016 Lin et al., 2016 Moorman et al., 2016). Small-worldness of MSC and SL were mostly above 1, but lower than 1 for ICOH, suggesting that functional EEG networks did not have small-world properties. As a result, statistically significant negative correlation occurred between FC and SW for all three FC measures. Three undirected FC measures were used: magnitude-squared coherence (MSC), imaginary part of coherency (ICOH), and synchronization likelihood (SL). For that purpose, Pearson correlation was calculated between FC and small-worldness (SW). The aim of the study was to analyze the relationship between resting state electroencephalographic (EEG) alpha functional connectivity (FC) and small-world organization. 2Department of Computer Systems, School of Information Technologies, Tallinn University of Technology, Tallinn, Estonia.1Centre for Biomedical Engineering, Department of Health Technologies, School of Information Technologies, Tallinn University of Technology, Tallinn, Estonia.Laura Päeske 1*, Hiie Hinrikus 1, Jaanus Lass 1, Jaan Raik 2 and Maie Bachmann 1
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |