search

OHBA-analysis / MEG-ROI-nets

Stationary network analysis for MEG
12 stars 10 forks source link

Brookes et al (2012) approach for leakage correction #7

Open suhnyoungjun opened 1 year ago

suhnyoungjun commented 1 year ago

Dear OBHA Team,

I appreciate all of the toolboxes kindly shared with the community. While Colclough's (2015) approach for source leakage correction has been widely used for the last few years, I'd like to preserve all the ROIs I have and am trying to use the Gram-Schmidt orthogonalization method that Brookes et al. (2012) used instead. Could you kindly share the codes you used for Brookes et al. (2012)?

Thank you for your help in advance!

Suhnyoung

woolrich commented 1 year ago

I assume you mean a pairwise correction? If so, this is not directly implemented. However, you could just use construct a loop over all pairs of ROIs and run the Colclough approach on each pair of ROIs to get a pairwise correction. This should be easy to do.

Note, however, that crucially the pairwise correction does not protect you from what we call "inherited", and others call "ghost" (see https://pubmed.ncbi.nlm.nih.gov/29477441/), artefactual interactions in the way that the multivariate Colclough approach does.

suhnyoungjun commented 1 year ago

Thank you so much for the suggestion! I really appreciate it. I will also keep in mind the possible artifactual interactions!