Centrality and interhemispheric coordination are related to different clinical/behavioral factors in attention deficit/hyperactivity disorder: a resting-state fMRI study

Abstract

Eigenvector-Centrality (EC) has shown promising results in the field of Psychiatry, with early results also pertaining to ADHD. Parallel efforts have focused on the description of aberrant interhemispheric coordination in ADHD, as measured by Voxel-Mirrored-Homotopic-Connectivity (VMHC), with early evidence of altered Resting-State fMRI. A sample was collected from the ADHD200-NYU initiative: 86 neurotypicals and 89 participants with ADHD between 7 and 18 years old were included after quality control for motion. After preprocessing, voxel-wise EC and VMHC values between diagnostic groups were compared, and network-level values from 15 functional networks extracted. Age, ADHD severity (Connor’s Parent Rating-Scale), IQ (Wechsler-Abbreviated-Scale), and right-hand dominance were correlated with EC/VMHC values in the whole sample and within groups, both at the voxel-wise and network-level. Motion was controlled by censoring time-points with Framewise-Displacement > 0.5 mm, as well as controlling for group differences in mean Framewise-Displacement values. EC was significantly higher in ADHD compared to neurotypicals in the left inferior Frontal lobe, Lingual gyri, Peri-Calcarine cortex, superior and middle Occipital lobes, right inferior Occipital lobe, right middle Temporal gyrus, Fusiform gyri, bilateral Cuneus, right Precuneus, and Cerebellum (FDR-corrected-p = 0.05). No differences were observed between groups in voxel-wise VMHC. EC was positively correlated with ADHD severity scores at the network level (at p-value < 0.01, Inattentive: Cerebellum rho = 0.273; Hyper/Impulsive: High-Visual Network rho = 0.242, Cerebellum rho = 0.273; Global Index Severity: High-Visual Network rho = 0.241, Cerebellum rho = 0.293). No differences were observed between groups for motion (p = 0.443). While EC was more related to ADHD psychopathology, VMHC was consistently and negatively correlated with age across all networks.

Reduced homotopic interhemispheric connectivity in psychiatric disorders: evidence for both transdiagnostic and disorder specific features

“…However, in medication-free children (8.5–9.1 years old) decreased homotopic rsFC of the MOG has been found based on a whole brain VMHC analysis with homotopic rsFC being negatively correlated with anxiety levels rated by parents (Zhou et al ., 2018 ). There is also one study reporting no homotopic rsFC changes in ADHD children/adolescents (7–18 years old; Tarchi et al ., 2022 ) (see Fig. 1 for summary).…”

Progressive Voxel‐Wise Homotopic Connectivity from childhood to adulthood: Age‐related functional asymmetry in resting‐state functional magnetic resonance imaging

“…In accordance with the original publication by Shirer et al. (2012), and with the literature using the template (Hall et al., 2013; Tarchi et al., 2022), the atlas was composed by the following networks: anterior salience, auditory, basal ganglia, dorsal and ventral default mode, language, left and right executive control, posterior salience, precuneus, sensorimotor, high and primary visual, and visuospatial. Mean VMHC values per network and their variance were calculated by averaging results per network mask.…”

“…Fourteen reference networks were obtained from the Functional Imaging in Neuropsychiatric Disorder Lab website-University of Stanford (Shirer et al, 2012). In accordance with the original publication by Shirer et al (2012), and with the literature using the template (Hall et al, 2013;Tarchi et al, 2022) Finally, as VMHC is a dimensionless variable, the degree of homotopy as measured by different scanning procedures could not be directly compared between the adult and minor samples.…”

Association between functional network connectivity, retina structure and microvasculature, and visual performance in patients after thalamic stroke: An exploratory multi‐modality study

“…Each of the brain regions (i.e., nodes) divided using AAL was associated with its corresponding resting-state functional network. Five empirical and robust functional networks (i.e., modules) (He et al, 2009) were further extracted from the resting-state function networks, consistent with previous research studies (Reber et al, 2021;Tarchi et al, 2022;Xin et al, 2022;Zhang et al, 2022), defined as follows: Sensorimotor network module, visual network module, attention network module, default mode network (DMN) module, and subcortical network module are detailed in Table S1 and Figure S1. The FC within each module was calculated as the average of all connection weights in the module.…”