Abstract
Brain functional disruption and cognitive shortfalls as consequences of neurodegeneration are among the most investigated aspects in current clinical research. Traditionally, specific anatomical and behavioral traits have been associated with neurodegeneration, thus directly translatable in clinical terms. However, these qualitative traits, do not account for the extensive information flow breakdown within the functional brain network that deeply affect cognitive skills. Behavioural variant Frontotemporal Dementia (bvFTD) is a neurodegenerative disorder characterized by behavioral and executive functions disturbances. Deviations from the physiological cognitive functioning can be accurately inferred and modeled from functional connectivity alterations. Although the need for unbiased metrics is still an open issue in imaging studies, the graph-theory approach applied to neuroimaging techniques is becoming popular in the study of brain dysfunction. In this work, we assessed the global connectivity and topological alterations among brain regions in bvFTD patients using a minimum spanning tree (MST) based analysis of resting state functional MRI (rs-fMRI) data. Whilst several graph theoretical methods require arbitrary criteria (including the choice of network construction thresholds and weight normalization methods), MST is an unambiguous modeling solution, ensuring accuracy, robustness, and reproducibility. MST networks of 116 regions of interest (ROIs) were built on wavelet correlation matrices, extracted from 41 bvFTD patients and 39 healthy controls (HC). We observed a global fragmentation of the functional network backbone with severe disruption of information-flow highways. Frontotemporal areas were less compact, more isolated, and concentrated in less integrated structures, respect to healthy subjects. Our results reflected such complex breakdown of the frontal and temporal areas at both intra-regional and long-range connections. Our findings highlighted that MST, in conjunction with rs-fMRI data, was an effective method for quantifying and detecting functional brain network impairments, leading to characteristic bvFTD cognitive, social, and executive functions disorders.
Highlights
The behavioral variant Frontotemporal Dementia (bvFTD) is clinically defined by personality changes and behavioral disturbances, impairment of executive functions and emotional blunting (Gorno-Tempini et al, 2011; Rascovsky et al, 2011)
We examined the complexity of resting state functional MRI (rs-fMRI) data (Saba et al, 2018) using bivariate measures computed as summary regional values of the wavelet correlations (Supplementary Table 2)
The significant increment of wavelet correlation diversity and its percentage of zeros indicated a decreased heterogeneity and increased null functional connectivity between brain regions in bvFTD compared to healthy controls (HC)
Summary
The bvFTD is clinically defined by personality changes and behavioral disturbances, impairment of executive functions and emotional blunting (Gorno-Tempini et al, 2011; Rascovsky et al, 2011). The increasing interest in unraveling functional and structural features of the brain, has benefitted from complex network analyses, such as graph theory, a multidisciplinary approach that allows to analyse complex systems in a straightforward computable way and to describe cerebral areas as nodes, and their connections as edges (Rubinov and Sporns, 2010). Conventional graph theoretical analyses are helpful in dissecting disease mechanisms (Bullmore and Sporns, 2012), the methodology is significantly hampered by a number of arbitrary choices Descriptive metrics and their normalization, network type (weighted or unweighted networks), threshold value (fixed cut-off, fixed average degree, fixed edge density, or variable threshold over a range of values) are some of the critical points making network results difficult to reproduce (van Wijk et al, 2010; Telesford et al, 2011; Stam et al, 2014; Drakesmith et al, 2015; Yu et al, 2016). Several network metrics and node centrality indices may assume different importance at either local or global scale (Telesford et al, 2011; Antonenko et al, 2018), whether the graph model accounts for time variant (i.e., dynamic) or invariant (i.e., static) connectivity (Rashid et al, 2014; Park et al, 2018), and the parcellation type, according to Independent Component Analysis (ICA; McKeown et al, 2003; Griffanti et al, 2014) and specific atlases (see Materials and Methods, for data pre-processing in this work)
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
