Characterisation of Haemodynamic Activity in Resting State Networks by Fractal Analysis

Abstract

Intrinsic brain activity is organized into large-scale networks displaying specific structural–functional architecture, known as resting-state networks (RSNs). RSNs reflect complex neurophysiological processes and interactions, and have a central role in distinct sensory and cognitive functions, making it crucial to understand and quantify their anatomical and functional properties. Fractal dimension (FD) provides a parsimonious way of summarizing self-similarity over different spatial and temporal scales but despite its suitability for functional magnetic resonance imaging (fMRI) signal analysis its ability to characterize and investigate RSNs is poorly understood. We used FD in a large sample of healthy participants to differentiate fMRI RSNs and examine how the FD property of RSNs is linked with their functional roles. We identified two clusters of RSNs, one mainly consisting of sensory networks (C1, including auditory, sensorimotor and visual networks) and the other more related to higher cognitive (HCN) functions (C2, including dorsal default mode network and fronto-parietal networks). These clusters were defined in a completely data-driven manner using hierarchical clustering, suggesting that quantification of Blood Oxygen Level Dependent (BOLD) signal complexity with FD is able to characterize meaningful physiological and functional variability. Understanding the mechanisms underlying functional RSNs, and developing tools to study their signal properties, is essential for assessing specific brain alterations and FD could potentially be used for the early detection and treatment of neurological disorders.

Hypothalamic structural integrity and temporal complexity of cortical information processing at rest in migraine without aura patients between attacks

“…The process was then repeated for every subject and every RSN. Higuchi's (FD) can be seen as a quantitative nonlinear measure of the BOLD signal dynamics 15 , 16 .…”

“…Therefore, we measure the hypothalamic microstructure through diffusion tensor imaging (DTI), a useful sensitive method to detect white matter tracts in grey matter nuclei-as is the case for the hypothalamus 10 and the independent cortical networking by acquiring resting-state functional MRI taking the advantages of the non-linearity of the Higuchi's fractal dimension (FD) analysis [11][12][13][14] . This non-linear approach is more suitable to describe the irregular and non-periodic patterns characterizing the BOLD signature of discrete cortical areas belonging to a resting-state network (RSN) recorded by neuroimaging 15,16 as well as electrophysiological techniques 17,18 .…”

“…www.nature.com/scientificreports/ for every subject and every RSN. Higuchi's (FD) can be seen as a quantitative nonlinear measure of the BOLD signal dynamics15,16 . As our primary endpoint was to detect differences in the hypothalamic microstructure between HC and MO, a sample size calculation was based on pilot data from 20 subjects, ten for each group, enrolled independently from the current study.…”

Haemodynamic activity characterization of resting state networks by fractal analysis and thalamocortical morphofunctional integrity in chronic migraine

“…These three measures were averaged to give one mean value of FD for each k, for each subject [14,20,38]. The process was then repeated for every subject and every RSN (in Porcaro and colleagues [39] it is shown in detail the procedure and additional analyses demonstrating that the FD measurements were not dependent on the choice of window length or overlapping windows).…”

Haemodynamic activity characterization of Resting State Networks by Fractal Analysis and thalamocortical morphofunctional integrity in chronic migraine

“…These three measures were averaged to give one mean value of FD for each k, for each subject [14,38,39]. The process was then repeated for every subject and every RSN (in Porcaro and colleagues [40] it is shown in detail the procedure and additional analyses demonstrating that the FD measurements were not dependent on the choice of window length or overlapping windows).…”