Hi Joelle,
I don't see anything wrong with your command.
However, if you want to email me directly the T1 and the example_func,
I can take a look for you. (No need to send the brain extracted image,
I can recreate it.)
On Mon, Aug 3, 2015 at 4:11 PM, Joelle Zimmermann
<[log in to unmask]> wrote:
> Unfortunately the size of the T1 file exceeds the fsl mailing list limit so
> I cant attach it here. I explain the error in more detail in my previous
> post. Please help!
> Thanks,
> Joelle
>
> On Mon, Aug 3, 2015 at 3:54 PM, Joelle Zimmermann
> <[log in to unmask]> wrote:
>>
>> I'm attaching the T1 image (this one is still uncorrected for NaN's).
>> Thanks a lot.
>>
>> On Mon, Aug 3, 2015 at 3:27 PM, Joelle Zimmermann
>> <[log in to unmask]> wrote:
>>>
>>> Thank you Mark and Niels.
>>>
>>> I tried rerunning fslmaths as you suggest Mark (fslmaths T1.nii -nan
>>> T1_nonan.nii), but am still getting the same error. These are the two
>>> commands I am running:
>>>
>>> bet ${path_to_anat_subj}/T1_nonan.nii ${path_to_anat_subj}/brain2.nii
>>> -R
>>>
>>> epi_reg --epi=${path_to_bold_subj}/featDir.feat/example_func.nii.gz
>>> --t1=${path_to_anat_subj}/T1_nonan.nii
>>> --t1brain=${path_to_anat_subj}/brain2.nii
>>> --out=${path_to_bold_subj}/featDir.feat/example_func2highres
>>>
>>> After which I get the error.
>>>
>>> I'll attach the T1 image shortly, just downloading from server.
>>>
>>> Thanks,
>>> Joelle
>>>
>>> On Mon, Aug 3, 2015 at 3:11 PM, Mark Jenkinson
>>> <[log in to unmask]> wrote:
>>>>
>>>> Hi,
>>>>
>>>> This isn't the correct usage of fslmaths - you need an input name,
>>>> followed by operations, followed by an output name. At the moment you have
>>>> not given an output name so you will not get an output. Do:
>>>> fslmaths T1 -nan T1_nonan
>>>> and then try and run things using the image T1_nonan
>>>>
>>>> Hopefully that will fix things.
>>>>
>>>> All the best,
>>>> Mark
>>>>
>>>>
>>>> From: FSL - FMRIB's Software Library <[log in to unmask]> on behalf of
>>>> Joelle Zimmermann <[log in to unmask]>
>>>> Reply-To: FSL - FMRIB's Software Library <[log in to unmask]>
>>>> Date: Monday, 3 August 2015 14:03
>>>> To: "[log in to unmask]" <[log in to unmask]>
>>>> Subject: Re: [FSL] set_bbr_seg ERROR
>>>>
>>>> Hi,
>>>>
>>>> Thanks again for this suggestion. A follow-up question here - I tried
>>>> using fslmaths to replace the NaN's by doing:
>>>>
>>>> fslmaths T1.nii -nan
>>>>
>>>> I'm not sure how to check if this actually did what I want it to do -
>>>> replace the NaN's with 0's; and if I am using the command correctly.
>>>>
>>>> After trying to correct the NaN's in the T1 image in this way, I run bet
>>>> , followed by epi_reg, and still get that same: ERROR:set_bbr_seg: could not
>>>> find any boundary points!
>>>>
>>>> Thanks,
>>>> Joelle
>>>>
>>>>
>>>> On Sat, Aug 1, 2015 at 1:52 PM, Mark Jenkinson
>>>> <[log in to unmask]> wrote:
>>>>>
>>>>> Hi,
>>>>>
>>>>> This error is almost always due to the fact that the segmentation of
>>>>> the structural image has failed. It might have just completely failed for
>>>>> some reason (including the job dying for various IT reasons like lack of
>>>>> disk space, or incorrect filename, as well as problems with the image such
>>>>> as strong artefact) or it could be caused if the image initially contained
>>>>> NaN values (Not a Number) due to it having been processed by some other
>>>>> package (e.g. matlab often inserts these). So check the structural image,
>>>>> check its segmentation, and check the structural image for NaN values (you
>>>>> can use the -nan or -nanm flag in fslmaths to fix/identify these).
>>>>>
>>>>> FYI: BBR is Boundary-Based Registration, and is the cost function used
>>>>> inside the registration method for EPI to structural registrations.
>>>>>
>>>>> All the best,
>>>>> Mark
>>>>>
>>>>>
>>>>> From: FSL - FMRIB's Software Library <[log in to unmask]> on behalf of
>>>>> Joelle Zimmermann <[log in to unmask]>
>>>>> Reply-To: FSL - FMRIB's Software Library <[log in to unmask]>
>>>>> Date: Saturday, 1 August 2015 00:38
>>>>> To: "[log in to unmask]" <[log in to unmask]>
>>>>> Subject: [FSL] set_bbr_seg ERROR
>>>>>
>>>>> Hi everyone,
>>>>>
>>>>> While registering one subject from functional>T1>standard space, I'm
>>>>> getting the following error:
>>>>>
>>>>> ERROR:set_bbr_seg: could not find any boundary points!
>>>>> WARNING: Found NaN in BBR cost(2)
>>>>>
>>>>> This is the case for only one of my 20 subjects. Does anybody have any
>>>>> ideas of why this happens; or where exactly the NaN's are found? (Not sure
>>>>> what a BBR cost is).
>>>>>
>>>>> Thanks for any suggestions,
>>>>> Joelle
>>>>
>>>>
>>>
>>
>
--
Niels Bergsland
Integration Director
Buffalo Neuroimaging Analysis Center
100 High St. Buffalo NY 14203
[log in to unmask]
