Hi Steve,
Thank you for your answer.
I apologize in advance as I have a follow-up question which I was hoping you can lend some help.
I followed your suggestion to use FLIRT and FNIRT with the -inweight and -inmask options, respectively, to mask out the lesion in stroke brains for registration. However, what are the steps I should take in order to input these registered brains for usage in Melodic?
In other words, if we were to conduct a seed-based analysis on resting state data using FEAT, this is not a problem since we can replace the registration output (in the “reg” folder) from the first-level FEAT analysis with the registered brains we want. Then, we can run a higher-level FEAT analysis since FEAT would retrieve the files from each subject’s first-level FEAT directory (which contain the registered brains we want).
However, in the case of Melodic, is there a way we can stop the Melodic GUI as it is running single-session ICA on each subject in order to replace the “reg” output with the registered brains we want and then continue the Melodic GUI for group-level ICA? Or perhaps, is there a way we can input the registered brains we want in the Melodic command line such that they can be used for multi-session temporal concatenation?
Thank you for any input you can provide!
Timothy
|