Hi Jen,
You should run the fsl course data through your version of feat
and check that you get sensible results. Just the av dataset (first
practical on feat) is enough to check. You can always do this in
the absence of feeds.
If this is fine but your data still has problems then upload the
offending
data to our upload site and we'll have a look.
All the best,
Mark
On 10 Sep 2008, at 23:00, Jennifer Bramen wrote:
> I used FEAT 4.0.4 for the analysis. I only see FEEDS 4.1 available
> online. Does anyone know where I can get a copy?
>
> Also, it seems like some analyses are working.
>
>
> On Sep 10, 2008, at 11:11 AM, Benny Liberg wrote:
>
>> Hi Jennifer,
>>
>> You might want to run the FEEDS evaluation package to see if FEAT is
>> working fine. Is other analyses working fine?
>>
>> Cheers,
>> Benny
>>
>> 2008/9/10 Jennifer Bramen <[log in to unmask]>:
>>> Right, I have seen the uncorrected maps, and they look fine. The
>>> corrected maps look totally bizarre given that the zstat images they
>>> are correcting have huge and intense clusters. I am happy to submit
>>> any relevant files (zstat, cope, varcope, mask, thresh_zstat, log)
>>> so
>>> we can get to the bottom of this. This is not the first post stat we
>>> have run on this data that looks bizzare and unbelievable, and I
>>> would
>>> very much like to figure out what the underlying issue is.
>>>
>>> Best regards
>>>
>>> Jennifer Bramen
>>> UCLA Laboratory of Neuroimaging
>>>
>>> On Sep 10, 2008, at 10:54 AM, Yoshiko Yamada wrote:
>>>
>>>> Hi Jennifer,
>>>>
>>>> Are you using the cluster thresholding option at post-stats? If so,
>>>> have you tried lowering the cluster p threshold? If you changed it
>>>> to "1", you should see the same maps as the uncorrected ones.
>>>>
>>>> -Yoshiko
>>>>
>>>>
>>>> On Wed, 10 Sep 2008 18:39:13 +0100, Jennifer Bramen
>>>> <[log in to unmask]
>>>>> wrote:
>>>>>
>>>>> Dear FSL List,
>>>>>
>>>>> We have been having a reoccurring problem with nonsensical post
>>>>> stats
>>>>> results. In one case, we are looking for the mean effect of task
>>>>> in
>>>>> a fairly
>>>>> large group of subjects (25). The uncorrected stats maps show very
>>>>> intense
>>>>> activation in many large clusters. However, the corrected maps
>>>>> show
>>>>> absolutely nothing. We thought perhaps it was a resource error,
>>>>> but
>>>>> we have
>>>>> rerun the analysis multiple times with no improvement. We also
>>>>> found no
>>>>> error in the log. Can someone suggest a course of action? I do not
>>>>> believe
>>>>> that these post stat results are accurate.
>>>>>
>>>>> Thank you in advance.
>>>>>
>>>>> Best regards
>>>>>
>>>>> Jennifer Bramen
>>>>> UCLA Laboratory of Neuroimaging
>>>>>
>>>
>>>
>>> IMPORTANT WARNING: This email (and any attachments) is only
>>> intended for
>>> the use of the person or entity to which it is addressed, and may
>>> contain
>>> information that is privileged and confidential. You, the
>>> recipient, are
>>> obligated to maintain it in a safe, secure and confidential manner.
>>> Unauthorized redisclosure or failure to maintain confidentiality may
>>> subject you to federal and state penalties. If you are not the
>>> intended
>>> recipient, please immediately notify us by return email, and delete
>>> this
>>> message from your computer.
>>>
>
>
> IMPORTANT WARNING: This email (and any attachments) is only
> intended for the use of the person or entity to which it is
> addressed, and may contain information that is privileged and
> confidential. You, the recipient, are obligated to maintain it in a
> safe, secure and confidential manner. Unauthorized redisclosure or
> failure to maintain confidentiality may subject you to federal and
> state penalties. If you are not the intended recipient, please
> immediately notify us by return email, and delete this message from
> your computer.
>
|