The ‘igamc underflow error’ isn’t a problem - this can happen naturally when parameters go out of bounds internally, but doesn’t normally mean the analysis fails (it is an error I have been unable to suppress thus far). I take it that it is working in some cases, but that you don’t have an obvious reason why some are going wrong? From memory we have sometimes had trouble where the mask has isolated voxels - so that might be worth checking, do you have log files to examine?
Michael
> On 1 Mar 2017, at 21:08, SUBSCRIBE FSL Ping-Hong Yeh <[log in to unmask]> wrote:
>
> Hello FSL users,
>
> Some of my attempts using verbena for DSC processing have failed; and i have noticed the "igamc underflow error" message shown on the terminal, but not sure if that is the cause. There is no outputs in the vm and vm_mv folders, but the modelfree results seem fine.
>
> Here is the command i used,
>
> verbena -i ${out_dir}/$sub/unwarp/unwarp.nii.gz -m ${out_dir}/$sub/unwarp/unwarpbrain_mask.nii.gz -a ${out_dir}/$sub/unwarp/aif.nii.gz -aifconc -tr 1.59 -te 0.0146 -modelfreeinit -mv -o ${out_dir}/$sub/unwarp
>
> , where the unwarp.nii.gz is the 4D dsc data.
>
>
> Thank you,
>
> Ping
|