[Freesurfer] SNR maps

Hi freesurfers,

I am using FSFAST on a group of 29 subjects...having run isxconcat-sess I am now doing the final quality control (Volume) step outlined in the FSFAST tutorial, which is to inspect the SNR maps. Inspecting the signal to noise ratio maps I am seeing 3 outliers (see attached), suggesting there is something strange going on with 3 participants' trials. I assume this means I should be excluding these participants from the analysis...but I am also wondering what the cause of the outliers might be, and whether it might be the result of a failed step earlier in the analysis. Is there anywhere else in the processing stream that I would be able to notice something is wrong?

Take it easy, -Christian