I want to test my imaging results generated from my deep learing method(.mat) with BST. Is there any way to implement that? Is it possible to just simply rewrite the parameters in a SOURCE FILE generated from a random method selected in BST?(such as LORETA)
the following are the parameters in a source file i created in BST database. Can you help me explain what these paremeters actually work in a source struct?
If I rewrite the ImageGridAmp with my matrix and keep the others unchanged though there might be some data dimension definition problem, will it work? thanks for your help!
Yes, you can do that. One thing you need to keep in mind is the difference between kernel and full sources. Check these two links:
https://neuroimage.usc.edu/brainstorm/Tutorials/SourceEstimation#Output_mode
https://neuroimage.usc.edu/brainstorm/Tutorials/SourceEstimation#On_the_hard_drive
In screenshot it seems the original file obtained with LORETA, is a kernel. So it has the field ImagingKernel with the imaging kernel ([nSources, nSensors]) and the field ImageGridAmp was empty. In your case, if the results of your source estimation method is full sources ([nSources, nTime]), you need to clear the field ImagingKernel and place the full sources in ImageGridAmp.
You could create a full sources file from a link source file, by using the process: Extract > Extract values
Thanks for your reply！It really helps me.
