Hi, the reason that just thresholding voxelwise the zstat image  
doesn't look like thresh_zstat is probably because you chose a  
different method than voxelwise thresholding in FEAT  ;-)

If you chose cluster-based thresholding then only clusters large  
enough to be significant will have been included in thresh_zstat - so  
that's not the same as just thresholding voxelwise in FSLView.

Cheers.


On 5 Mar 2009, at 00:57, Lin Nga wrote:

> Liam
>
> Typing 2.3 into the min box works just fine. My question is just,  
> why is it
> that for the same given range, my zstat does not look like my  
> thresh_zstat
> image? From my understanding, thresh_zstat is the thresholded z  
> statistics
> image and zstat is the raw z statistics image; is this wrong? If I  
> play
> around with the threshold with the zstat image, does this not tell me
> anything significant?
>
> Thanks again,
> Lin
>
> On Wed, 4 Mar 2009 16:40:34 -0800, Nestor, Liam (Contractor)
> <[log in to unmask]> wrote:
>
>> Lin
>>
>> Overlay the thresh_zstat. When you do this, your underlay image  
>> should be
> entirely covered by (but you should able to see signs of clusters
> underneath). Now type 2.3 into the min box. Does typing 2.3 (your z  
> stat
> threshold as used in FEAT) still leave the brain very covered in  
> activity?
> If so, playing around with the min value (e.g., try typing 2.81  
> which is
> p=0.005) might give you a better idea of thresholding for an
> additiona/futurel FEAT analysis.
>>
>> Your axial (horizontal) image map in fslview should look like that  
>> in the
> report. I have just checked some of my files to confirm this.
>>
>> Liam.
>>
>>
>>
>>
>> *******************************************
>> Liam Nestor, Ph.D
>> Office C8-523
>> Laboratory for Molecular Neuroimaging
>> Semel Institute for Neuroscience and Human Behavior
>> 760 Westwood Plaza
>> Los Angeles 90024
>> Tel: 310-206-0655
>> E-mail: [log in to unmask]
>> *******************************************
>> ________________________________________
>> From: FSL - FMRIB's Software Library [[log in to unmask]] On Behalf  
>> Of Lin
> Nga [[log in to unmask]]
>> Sent: Wednesday, March 04, 2009 4:13 PM
>> To: [log in to unmask]
>> Subject: Re: [FSL] Discrepancies between feat output and fslview
>>
>> Hi Liam,
>>
>> Thank you for your response. Typing 2.3 into the min box does  
>> indeed make
>> the image look like the feat output but that wasn't really my main
>> confusion. What I'd like to know is that if I overlay the raw zstat  
>> file and
>> change the min and max threshold to reflect those given in the feat  
>> report (
>> 2.3 and 3.4 respectively ), this does not look anything like the feat
>> report. If I want to play around with the threshold to see if this  
>> affects
>> my results, should I not be doing so with the zstatX.nii.gz and use
>> thresh_zstatX.nii.gz instead? But isn't the thresh_zstatX.nii.gz  
>> image a
>> result of thresholding at 2.3?
>>
>> Thanks,
>> Lin
>>
>> On Wed, 4 Mar 2009 14:49:12 -0800, Nestor, Liam (Contractor)
>> <[log in to unmask]> wrote:
>>
>>> Lin
>>>
>>> I think what is happening here is you are thresholding at 2.3 in  
>>> your FEAT
>> analyses; the end product of which is a FEAT report telling you  
>> that the
>> range is from 2.3 (min) to 3.4 (max).
>>>
>>> When you overlay your thresholded zstat file on top of your  
>>> 152MNI, you
>> need to type 2.3 into the min box provided.
>>>
>>> Try doing this and see what happens.
>>>
>>> Liam.
>>>
>>> *******************************************
>>> Liam Nestor, Ph.D
>>> Office C8-523
>>> Laboratory for Molecular Neuroimaging
>>> Semel Institute for Neuroscience and Human Behavior
>>> 760 Westwood Plaza
>>> Los Angeles 90024
>>> Tel: 310-206-0655
>>> E-mail: [log in to unmask]
>>> *******************************************
>>> ________________________________________
>>> From: FSL - FMRIB's Software Library [[log in to unmask]] On  
>>> Behalf Of Lin
>> Nga [[log in to unmask]]
>>> Sent: Wednesday, March 04, 2009 2:47 PM
>>> To: [log in to unmask]
>>> Subject: Re: [FSL] Discrepancies between feat output and fslview
>>>
>>> I also want to add a couple more thing I noted. I opened up the  
>>> thresh_zstat
>>> in fslview, which should look like what is displayed on the feat  
>>> report (and
>>> it does as expected), and the threshold ranges from 0 to 2.9. Why  
>>> does the
>>> range on the feat report say 2.3 to 3.4 when the thresholded image  
>>> goes from
>>> 0 to 2.9? Could someone please explain what is going on here?
>>>
>>> Thanks for the help,
>>> Lin
>>>
>>> On Wed, 4 Mar 2009 21:38:13 +0000, Lin Nga <[log in to unmask]> wrote:
>>>
>>>> Hi everyone,
>>>>
>>>> I am looking at one zstat image for one cope of a group analysis.  
>>>> I have the
>>>> zstat overlay the bg_image in fslview and I adjusted the  
>>>> threshold to match
>>>> that of the feat output (in my case 2.3 to 3.4). What I see in  
>>>> fslview does
>>>> not look like what I see in the feat output. What is wrong here?  
>>>> Shouldn'
>>>> the two look the same? Am I looking at the wrong statistical image?
>>>>
>>>> Thank you so much,
>>>> Lin
>>>
>>> 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.
>


---------------------------------------------------------------------------
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
---------------------------------------------------------------------------