Power spectrum of resting-state blood-oxygen-level-dependent signal
Authors:
- 29 August 2019
- journal article
- research article
- Published by American Physical Society (APS) in Physical Review E
- Vol. 100 (2) , 022418
Abstract
Hemodynamic modeling is used to explore the origin, predict, and analyze the power spectrum of the resting-state blood-oxygen-level-dependent (BOLD) signal measured by functional magnetic resonance imaging (fMRI), which has been reported to have a power-law form, i.e., , where is the power, is the frequency, and is the power-law exponent. However, current fMRI experimental paradigms have limited acquisition durations, affecting the spectral resolution of fMRI data at the low-frequency regime. Here, the claimed power-law spectrum is investigated by using a recent hemodynamic model to analytically derive the BOLD power spectrum, with parameters that are related to neurophysiology. The theoretical results show that, for all realistic parameter combinations, the BOLD power spectrum is flat at , has a weak resonance originating from intrinsic oscillations of vasodilatory response, and becomes a power law for high frequencies, all of which is in agreement with an empirical data set that describes the spectrum of one subject and brain region. However, the results are contrary to studies reporting a pure power-law spectrum at . The discrepancy is attributed largely to data averaging employed by current approaches that averages together important properties of the BOLD power spectrum, such as its resonance, that biases the spectrum to only show a power law. Data averaging also reduces the high-frequency power-law exponent relative to individual cases. Overall, this work demonstrates how the model can reproduce BOLD dynamics and further analyze its low-frequency behavior. Moreover, it also uses the model to explain the impact of procedures, such as data averaging, on the reported features of the BOLD power spectrum.
Keywords
Funding Information
- Centre of Excellence for Electromaterials Science, Australian Research Council (CE140100007)
- Australian Research Council (FL140100025, DP170101778)
This publication has 65 references indexed in Scilit:
- Altered power spectral density in the resting-state sensorimotor network in patients with myotonic dystrophy type 1Scientific Reports, 2018
- Spontaneous Neural Dynamics and Multi-scale Network OrganizationFrontiers in Systems Neuroscience, 2016
- Studying Brain Organization via Spontaneous fMRI SignalNeuron, 2014
- Twenty years of functional MRI: The science and the storiesNeuroImage, 2012
- What we can do and what we cannot do with fMRINature, 2008
- The power of spectral density analysis for mapping endogenous BOLD signal fluctuationsHuman Brain Mapping, 2008
- Spontaneous fluctuations in brain activity observed with functional magnetic resonance imagingNature Reviews Neuroscience, 2007
- BOLD responses to stimuli: Dependence on frequency, stimulus form, amplitude, and repetition rateNeuroImage, 2006
- Empirical Analyses of BOLD fMRI StatisticsNeuroImage, 1997
- Functional connectivity in the motor cortex of resting human brain using echo‐planar mriMagnetic Resonance in Medicine, 1995