How to plot t-maps on surface

AntonioMaffei · August 17, 2019, 12:17pm

Dear all,

I have computed a series of analyses, with custom MATLAB code, on the 68 time series extracted from the source-reconstructed activity after downsampling with the Desikan atlas. These analyses resulted in a set of t-values for each ROI.

Now, I would like to go back in Brainstorm and plot these t-values over a cortical surface in order to obtain a t-map of cortical activity, but I am struggling in understanding what is the correct approach to accomplish this task.

Any suggestion will be greatly appreciated!

Antonio

Francois · August 21, 2019, 2:42pm

If you want map your 68 values back to a surface with 15000 vertices, you need to create manually a source file, as done in the process "Simulate recordings from scouts":

github.com

brainstorm-tools/brainstorm3/blob/master/toolbox/process/functions/process_simulate_recordings.m#L156-L196


      
          else

              % Check head model

              if isempty(sStudyChannel.iHeadModel)

                  bst_report('Error', sProcess, [], ['No head model file available.' 10 'Please calculate a head model before running simulations.']);

                  return;

              end

              % Load headmodel

              HeadModelFile = sStudyChannel.HeadModel(sStudyChannel.iHeadModel).FileName;

              HeadModelMat = in_bst_headmodel(HeadModelFile);

          end

          % If no orientations: error

          if isempty(HeadModelMat.GridOrient)

              bst_report('Error', sProcess, [], 'No source orientations available in this head model.');

              return;

          end

          % Get all the MEG/EEG channels

          Modalities = {};

          if ~isempty(HeadModelMat.MEGMethod)

              Modalities{end+1} = 'MEG';

          end

This file has been truncated. show original

This procedure is also discussed on this forum, for example:

AntonioMaffei · August 21, 2019, 3:40pm

Hi Francois,

thanks for your reply. I have found also another solution, that is downsampling to the Desikan Atlas a source file from my protocol and then replacing the ImageGridAmp field with my own t-values and reimporting the file into Brainstorm.

Apart from the unit scale, your suggested solution and mine give both the image that I wanted to create .

Thanks!