Exporting cortical activity to Matlab

[NingZhuang]

Exporting Source Activations to Matlab

Hi Tim, Indeed, the syntax of tess_smooth_sources changed and is documented in the very code of the function: W = tess_smooth_sources(Vertices, Faces, VertConn=[], FWHM=0.010, Method=‘average’) This function returns a smoothing kernel W, which is a linear operator [Nsources x Nsources] that you apply on your source matrix to smooth them over the surface. It does not estimate anything. Cheers, Francois

Hi, Francois,

Brainstrom can locate the source using dSPM and sLoreta algorithms. Now I have completed the process of source estimation.

For a specific subject, I input ERP signals, a matrix of 61 * 2816 (channel * time).

I want to take out the reconstructed cortical activity and do further causal network analysis (using the SIFT toolbox). Cortical activity refers to the reconstructed time series on the cortex, which should also be a matrix of 61*2816 (Am I right?).

In the "brainstorm_db" folder, which .mat file corresponds to the reconstructed cortical activity.

The input file is e1_Mean_ERP， a matrix with dimension 61*2816.

(1) Is the reconstructed cortical activity in "data_e1_Mean_ERP" , as shown in Fig.1?

(2) After opening the data_e1_Mean_ERP file in matlab, “F” is a matrix with dimension 61*2816, as shown in Fig.2. Is "F" the reconstructed cortical time series?

I look forward to your reply.

Ning

[Francois]

Cortical activity refers to the reconstructed time series on the cortex, which should also be a matrix of 61*2816 (Am I right?).

No, it corresponds to something probably much larger than this: number of dipoles in your source space x time samples.

(1) Is the reconstructed cortical activity in "data_e1_Mean_ERP" , as shown in Fig.1?

No, these are your averaged EEG recordings. The reconstructed source activity is in the files starting with the tag "results", in the variable "ImageGridAmp":
https://neuroimage.usc.edu/brainstorm/Tutorials/SourceEstimation#On_the_hard_drive

If you haven't read them yet, I would recommend you read and follow all the introduction tutorials, at least until #22:
https://neuroimage.usc.edu/brainstorm/Tutorials

Note that in general, we recommend importing the continuous recordings (at least the individual trials) and do the averaging in Brainstorm, in order to be able to build correct noise statistics for the source estimation.

[NingZhuang]

Thank you very much for your reply. I will import the original multi-trial EEG signal and calculate the noise covariance in brainstrom.