Poster No:
1784
Submission Type:
Abstract Submission
Authors:
Kimberly Rogge-Obando1, Caroline Martin1, Terra Lee1, Sarah Goodale1, Shiyu Wang1, Ruogi Yang1, Richard Song1, Jeremy Hogeveen2, Catie Chang1
Institutions:
1Vanderbilt University, Nashville, TN, 2University of New Mexico, Albuquerque, NM
First Author:
Co-Author(s):
Terra Lee
Vanderbilt University
Nashville, TN
Introduction:
The functional connectivity between anxiety and the default mode, central executive, and salience networks have been well established.[1,2] However, full-brain fluctuations that stem from arousal states and physiological influences such as heart rate have also been shown to influence functional connectivity. This has led a portion of the field choosing to remove these influences using global signal regression (whole-brain average BOLD signal). However, recent studies indicate that the global signal holds valuable information about psychiatric disorders.[3] Here, we determine the amount and extent to which regressing out the global signal, related measures of arousal, and heart rate impacts the relationship between functional connectivity and measures of anxiety. Additionally, we examine the spatial distribution of these global effects correlated to anxiety across the brain.
Methods:
Resting-state fMRI and psychometrics data from the Nathan Kline Institute (NKI-Rockland Sample,[4] was analyzed for this study. A total of 481 subjects that had State and Trait Anxiety Inventory scores greater than 30 were included in the analysis (F=328, M=153). FSL MELODIC ICA was used to derive five networks of interest: ventral-default mode (VDMN), dorsal-default mode (DDMN), salience (SAL), left-central executive (LCEN), and right-central executive (RCEN). Dual regression was used to extract subject-specific time series of each network. After motion regression, functional connectivity was computed between all pairs of networks and correlated to psychometrics, covarying for age and gender. This analysis was repeated after separately regressing out the global signal, a low-frequency heart rate signal, and an fMRI-derived arousal signal.[5] Lastly, spatial maps reflecting the association of global signal, arousal, and heart rate variability to fMRI signals across the brain were also constructed. FSL randomise was used to relate these spatial maps to anxiety measures while covarying for age, gender, and ethnicity and correcting for multiple comparisons.
Results:
Before global signal regression, the connectivity between the VDMN-SAL and SAL-RCEN was significant for state anxiety and SAL-RCEN connectivity was significant for trait anxiety. The regression of global signal and full-brain arousal fluctuations removed any significant network connections. However, when regressing heart rate related fluctuations, these functional connectivity values remained significant. For our arousal spatial maps, areas of the para hippocampal, intra parietal lobe and visual cortex were related to state anxiety but not to trait anxiety. There was no relationship between anxiety measures and either global signal maps or heart rate variation maps.
Conclusions:
Our results show that global signal regression impacted the significance of resting-state network functional connectivity and their relationship with anxiety. These results were replicated when regressing out an fMRI-derived arousal metric but not for heart rate. These results suggest that global fMRI signals, including arousal effects, may hold valuable information related to anxiety. This notion is further supported by our observation that key regions of the brain demonstrated arousal-related BOLD signals that were related to state anxiety.
Disorders of the Nervous System:
Psychiatric (eg. Depression, Anxiety, Schizophrenia) 2
Modeling and Analysis Methods:
fMRI Connectivity and Network Modeling 1
Keywords:
Anxiety
Computational Neuroscience
Data analysis
Design and Analysis
FUNCTIONAL MRI
1|2Indicates the priority used for review
Provide references using author date format
Saviola, F. et al. (2020),' Trait and state anxiety are mapped differently in the human brain', Scientific Reports 2020
Menon, V. (2011),' Large-scale brain networks and psychopathology: a unifying triple network model' Trends Cogn Sci
Hahamy, A. et al.(2014)' Save the Global: Global Signal Connectivity as a Tool for Studying Clinical Populations with Functional Magnetic Resonance Imaging', Brain Connect
Tobe, R. H. et al.(2022), 'A longitudinal resource for studying connectome development and its psychiatric associations during childhood',Scientific Data
Goodale, S. E. et al.(2021) 'Fmri-based detection of alertness predicts behavioral response variability', Elife