[Paper Review] Neurovascular coupling: insights from multi-modal dynamic causal modelling of fMRI and MEG

Amirhossein Jafarian, Vladimir Litvak+3 more|arXiv (Cornell University)|Mar 18, 2019

Functional Brain Connectivity Studies32 references5 citations

TL;DR

This paper proposes a multi-modal dynamic causal modeling (DCM) framework that integrates fMRI and MEG data to investigate neurovascular coupling mechanisms in the human brain. By using fMRI to localize neuronal sources and informing MEG DCM with these priors, the method estimates neuronal drive functions and models their influence on BOLD responses, revealing that the BOLD signal is instantaneously mediated by intrinsic neuronal dynamics and that neurovascular coupling is region-specific.

ABSTRACT

This technical note presents a framework for investigating the underlying mechanisms of neurovascular coupling in the human brain using multi-modal magnetoencephalography (MEG) and functional magnetic resonance (fMRI) neuroimaging data. This amounts to estimating the evidence for several biologically informed models of neurovascular coupling using variational Bayesian methods and selecting the most plausible explanation using Bayesian model comparison. First, fMRI data is used to localise active neuronal sources. The coordinates of neuronal sources are then used as priors in the specification of a DCM for MEG, in order to estimate the underlying generators of the electrophysiological responses. The ensuing estimates of neuronal parameters are used to generate neuronal drive functions, which model the pre or post synaptic responses to each experimental condition in the fMRI paradigm. These functions form the input to a model of neurovascular coupling, the parameters of which are estimated from the fMRI data. This establishes a Bayesian fusion technique that characterises the BOLD response - asking, for example, whether instantaneous or delayed pre or post synaptic signals mediate haemodynamic responses. Bayesian model comparison is used to identify the most plausible hypotheses about the causes of the multimodal data. We illustrate this procedure by comparing a set of models of a single-subject auditory fMRI and MEG dataset. Our exemplar analysis suggests that the origin of the BOLD signal is mediated instantaneously by intrinsic neuronal dynamics and that neurovascular coupling mechanisms are region-specific. The code and example dataset associated with this technical note are available through the statistical parametric mapping (SPM) software package.

Motivation & Objective

• To develop a Bayesian fusion framework integrating fMRI and MEG data to model neurovascular coupling.
• To identify the neuronal signals—pre- or post-synaptic—that drive the BOLD response.
• To determine whether neurovascular coupling is instantaneous or delayed and varies across brain regions.
• To apply variational Bayesian inference and Bayesian model comparison to select the most plausible neurovascular coupling model.
• To provide a reproducible method using SPM software with open-access code and data.

Proposed method

• fMRI data is used to localize neuronal sources, which inform the spatial priors in a DCM for MEG.
• MEG DCM estimates underlying neuronal generators using the fMRI-derived source coordinates.
• Neuronal parameters from MEG are used to generate drive functions representing pre- or post-synaptic activity under experimental conditions.
• These drive functions serve as inputs to a neurovascular coupling model estimated from fMRI data.
• Variational Bayesian inference is used to estimate model parameters, and Bayesian model comparison selects the most plausible model.
• The framework enables testing of biologically informed hypotheses about the origin of the BOLD response.

Experimental results

Research questions

• RQ1 Which neuronal signal—pre- or post-synaptic—most plausibly drives the BOLD response in the human brain?
• RQ2 Is neurovascular coupling instantaneous or delayed in its effect on hemodynamic responses?
• RQ3 Does the mechanism of neurovascular coupling vary across different brain regions?
• RQ4 Which model of neurovascular coupling best explains the multimodal fMRI and MEG data?
• RQ5 Can a Bayesian fusion of fMRI and MEG data improve the identification of neuronal sources and their hemodynamic consequences?

Key findings

• The BOLD signal is most plausibly explained by an instantaneous mediation of intrinsic neuronal dynamics.
• Neurovascular coupling mechanisms are region-specific, indicating variability in the neurovascular response across cortical areas.
• The most supported model features pre- or post-synaptic neuronal activity as the primary driver of the BOLD response.
• Bayesian model comparison successfully identified the best-fitting model among a set of biologically plausible hypotheses.
• The framework enables precise estimation of neuronal drive and hemodynamic responses using multimodal data fusion.
• The method is implemented and made available via the SPM software package with open-source code and example data.