Limits...
Construct validation of a DCM for resting state fMRI.

Razi A, Kahan J, Rees G, Friston KJ - Neuroimage (2014)

Bottom Line: Dynamic causal modelling (DCM) is a framework that allows for the identification of the causal (directed) connections among neuronal systems--known as effective connectivity.We also simulated group differences and compared the ability of spectral and stochastic DCMs to identify these differences.We show that spectral DCM was not only more accurate but also more sensitive to group differences.

View Article: PubMed Central - PubMed

Affiliation: The Wellcome Trust Centre for Neuroimaging, University College London, 12 Queen Square, London WC1N 3BG, UK; Department of Electronic Engineering, NED University of Engineering and Technology, Karachi, Pakistan. Electronic address: a.razi@ucl.ac.uk.

Show MeSH

Related in: MedlinePlus

This figure summarises the results of model inversion using the model and data of previous figure. We only show the results for connections with non-trivial connection strengths greater than 0.5 Hz (and omit self-connections for simplicity). The upper row shows the results for the spectral DCM in same format in previous figures for simulated data. The leftmost panel shows the Bayesian parametric averages over 22 subjects. The middle panel shows the results of classical t-tests reporting t-statistics for each connection, whereas the right panel shows only those edges on the graph that survive the corrected threshold in the middle panel. The lower row reports the results for stochastic DCM in the same format as for the spectral DCM.
© Copyright Policy - CC BY
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4295921&req=5

f0050: This figure summarises the results of model inversion using the model and data of previous figure. We only show the results for connections with non-trivial connection strengths greater than 0.5 Hz (and omit self-connections for simplicity). The upper row shows the results for the spectral DCM in same format in previous figures for simulated data. The leftmost panel shows the Bayesian parametric averages over 22 subjects. The middle panel shows the results of classical t-tests reporting t-statistics for each connection, whereas the right panel shows only those edges on the graph that survive the corrected threshold in the middle panel. The lower row reports the results for stochastic DCM in the same format as for the spectral DCM.

Mentions: Post-hoc model optimization found the fully connected model to have the largest free energy, consistent with previous similar analyses (Li et al., 2012). Bayesian parameter averaging (BPA) was then used to calculate the expected posterior connectivity estimates, and posterior confidence intervals. The concordance between stochastic and spectral results was subsequently examined qualitatively. In Fig. 10, we show the results of both the stochastic and spectral DCM schemes. The left columns show the results of BPA. To focus on non-trivial connections, we only report connection strengths that exceed 0.05 Hz in strength. We also omit self-connections in these plots for simplicity. BPA results are fairly consistent when compared between the two schemes, especially the connections originating from the bilateral intraparietal cortex. There is some disagreement in some other connections originating from the PCC and mPFC. Furthermore, coupling strengths from stochastic DCM are relatively smaller in magnitude than those from the spectral DCM, which is in accordance with previous simulations demonstrating that stochastic DCM tends to underestimate the effective connectivity (see also above). In the middle column, we show the results of classical t-tests to see which of the connections are significantly different from zero. In the last column, only the significant connections surviving the corrected threshold (the broken red line) are shown.


Construct validation of a DCM for resting state fMRI.

Razi A, Kahan J, Rees G, Friston KJ - Neuroimage (2014)

This figure summarises the results of model inversion using the model and data of previous figure. We only show the results for connections with non-trivial connection strengths greater than 0.5 Hz (and omit self-connections for simplicity). The upper row shows the results for the spectral DCM in same format in previous figures for simulated data. The leftmost panel shows the Bayesian parametric averages over 22 subjects. The middle panel shows the results of classical t-tests reporting t-statistics for each connection, whereas the right panel shows only those edges on the graph that survive the corrected threshold in the middle panel. The lower row reports the results for stochastic DCM in the same format as for the spectral DCM.
© Copyright Policy - CC BY
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4295921&req=5

f0050: This figure summarises the results of model inversion using the model and data of previous figure. We only show the results for connections with non-trivial connection strengths greater than 0.5 Hz (and omit self-connections for simplicity). The upper row shows the results for the spectral DCM in same format in previous figures for simulated data. The leftmost panel shows the Bayesian parametric averages over 22 subjects. The middle panel shows the results of classical t-tests reporting t-statistics for each connection, whereas the right panel shows only those edges on the graph that survive the corrected threshold in the middle panel. The lower row reports the results for stochastic DCM in the same format as for the spectral DCM.
Mentions: Post-hoc model optimization found the fully connected model to have the largest free energy, consistent with previous similar analyses (Li et al., 2012). Bayesian parameter averaging (BPA) was then used to calculate the expected posterior connectivity estimates, and posterior confidence intervals. The concordance between stochastic and spectral results was subsequently examined qualitatively. In Fig. 10, we show the results of both the stochastic and spectral DCM schemes. The left columns show the results of BPA. To focus on non-trivial connections, we only report connection strengths that exceed 0.05 Hz in strength. We also omit self-connections in these plots for simplicity. BPA results are fairly consistent when compared between the two schemes, especially the connections originating from the bilateral intraparietal cortex. There is some disagreement in some other connections originating from the PCC and mPFC. Furthermore, coupling strengths from stochastic DCM are relatively smaller in magnitude than those from the spectral DCM, which is in accordance with previous simulations demonstrating that stochastic DCM tends to underestimate the effective connectivity (see also above). In the middle column, we show the results of classical t-tests to see which of the connections are significantly different from zero. In the last column, only the significant connections surviving the corrected threshold (the broken red line) are shown.

Bottom Line: Dynamic causal modelling (DCM) is a framework that allows for the identification of the causal (directed) connections among neuronal systems--known as effective connectivity.We also simulated group differences and compared the ability of spectral and stochastic DCMs to identify these differences.We show that spectral DCM was not only more accurate but also more sensitive to group differences.

View Article: PubMed Central - PubMed

Affiliation: The Wellcome Trust Centre for Neuroimaging, University College London, 12 Queen Square, London WC1N 3BG, UK; Department of Electronic Engineering, NED University of Engineering and Technology, Karachi, Pakistan. Electronic address: a.razi@ucl.ac.uk.

Show MeSH
Related in: MedlinePlus