Hi,
Just as a note, the only difference between all structure and single
structures (apart form convenience) is in the boundary correction. The
surfaces output from FIRST are exactly the same.
'-m fast ' options is suggested, though I think this is the default. It
uses fast within an ROI surrounding the structure to perform the boundary
correction, eliminating the need for an arbitrary threshold.
Are you sure you have the most up-to-date version? What output do you get
if you type 'run_first_all' with no arguments
cheers,
brian
>> Hello,
> I am not sure my previous response was received (regarding missing output
> following the run_first_all command), so I am re-sending it:
>
>
> I ran FIRST without specifying structures, and it gave me output for all
> structures (I am interested in many subcort structures, so running a
> single structure is of no interest to me). I think my confusion has to do
> with the boundary correction method. When I type
>
> run_first_all <input file.nii.gz> <output file.nii.gz>
>
> I am prompted for a threshold value. Yet based on the FIRST information
> page, the run_first_all command should assume a standard boundary
> correction (which yields the files I am missing, _firstseg.nii.gz and
> _origsegs.nii.gz). I cannot run the command without specifying a
> threshold.
>
> There are other odd aspects to the FIRST commands. For example, when I add
> the arguments -i and -o, it does not execute the run_first_all command;
> when I leave those out, it does run them. And I am quite sure I'm using
> the right page, since it has the information on single structures that
> Mark updated last week (yet the title of the page is FIRST v1.2, not 1.4
> as I would expect...).
>
> Thanks,
> Melissa
>
> ________________________________________
> From: FSL - FMRIB's Software Library [[log in to unmask]] On Behalf Of
> Mark Jenkinson [[log in to unmask]]
> Sent: Friday, July 10, 2009 9:29 AM
> To: [log in to unmask]
> Subject: Re: [FSL] FIRST tie-up
>
> Hi,
>
> Did you specify a single structure or multiple structures?
> If you used a single structure then the naming is slightly
> different, as I just posted to a previous query on the
> mailing list a few minutes ago. We have updated our
> documentation to be explicit about this now.
>
> As for th4 vs fast, did you specify the thresholding
> method for boundary correction? That would give you
> an output like this. Also, are you using the latest method
> (FSL4.1.4) as the naming has changed since the previous
> method.
>
> Hope this helps.
> All the best,
> Mark
>
>
> On 10 Jul 2009, at 00:48, Melissa wrote:
>
>> I have just tried to play around with FIRST but original command
>> run_first_all
>> does not produce each type of output listed on the FSL FIRST web page.
>> Specifically, I see no
>>
>> _all_fast_firstseg.nii.gz
>> nor
>> _all_fast_origsegs.nii.gz
>>
>> instead I have the suffix _all_th4_first.nii.gz
>>
>> output. I followed all of the directory maintenance guidelines.
>> Any ideas of
>> where I have gone wrong?
>>
>> thanks,
>> Melissa
>>
>
|