Hi,
Try passing the data through fslmaths first - as this happens in prestats it's hit nothing to do with melodic per se but fails to get processed at a much lower level. You can cut&paste the commands from the log file one at a time in a terminal and trace what's going wrong that way...
Hth
Christian
Sent from my iPad
On 07.12.2012, at 20:00, Barry Bohnet <[log in to unmask]> wrote:
> The gradient echo header is the exact same as the spin echo header, except
> for a difference in pixdim7, but that differs between some of the gradient
> echo scans and they work, so I don't think that's the problem. I opened up
> the images in fslview and the image is clearly visible, though I would say
> that the spin-echo images do have a slighlty lower mean intensity than the
> gradient-echo images.
>
> I managed to get the spin-echo images to load, but I have to delete 25 (of
> 100) volumes, though it didn't work when I deleted 0,2,3,4,6,8 or 10
> volumes. Even with the gradient-echo images I couldn't get the program to
> run to completion if I deleted 0, or 8 volumes, but it would work if I
> deleted 3 or 10 volumes. I thought it might have something to do with the
> intensity of the first volume used, but I checked the intensity of all the
> volumes and they're consistent enough that that shouldn't be the issue.
> Why would there be problems depending on the number of volumes deleted?
>
> As always, any help is greatly appreciated. Thanks!
>
>
> Barry
>
>
>
>> Hi - I'm guessing that there's something weird/wrong with your input data.
>> What does fslhd report on the spin echo input data? Have you looked at
>> it in FSLview? Does it have a sensible (not too close to zero) mean
>> intensity level in the brain?
>>
>> cheers.
>>
>>
>> On 6 Dec 2012, at 22:36, Barry Bohnet wrote:
>>
>>> Hi there,
>>>
>>> I'm trying to run Melodic ICA on some EPI data-sets but I seem to keep
>>> running into some problems. The scans are 4-shot EPI, TR=0.5s, and I've
>>> set the High Pass Filter cutoff to 4 seconds and deleted 4 volumes.
>>> These
>>> settings seem to work for the gradient echo scans, but not the spin echo
>>> scans for some reason.
>>>
>>> In the log there's an issue under 'Pre-Stats' that says:
>>>
>>> "Cannot open volume prefiltered_func_data_smooth -mas mask
>>> prefiltered_func_data_smooth for reading!
>>> ...
>>> (file "usr/local/fsl/bin/feat" line 119)"
>>>
>>> This seems to be the same error if I delete 8 volumes. I've attatched
>>> the
>>> fslerrorreport from the described scan.
>>>
>>> Any help would be appreciated. Thanks!
>>>
>>>
>>> Barry
>>> <fsl_fCjSgi.gz>
>>
>>
>> ---------------------------------------------------------------------------
>> Stephen M. Smith, Professor of Biomedical Engineering
>> Associate Director, Oxford University FMRIB Centre
>>
>> FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK
>> +44 (0) 1865 222726 (fax 222717)
>> [log in to unmask] http://www.fmrib.ox.ac.uk/~steve
>> ---------------------------------------------------------------------------
>>
>>
>>
>>
|