Hi!
Ivana:
Thanks for the advices. I checed again the maximum values in the
activation volumes I realized that they are approximately 100 000 times
too large (maybe from s -> ms change or something). I quess the values
should be in seconds in the volumes even the MRpar matrix is in ms.
Where in fsl website can I find this patched version of POSSUM?
At the moment the only new file in patches folder,
http://www.fmrib.ox.ac.uk/fsldownloads/patches/, is that "mhanke.tar.gz"
file.
Mark:
The activation data is derived from fMRI data according the equation S =
S0 exp(-TE/T2* ) as you explained in earlier mails. I need to check
these computations again to figure out how the values were so large.
With the smaller values the simulation looks to work fine.
Thanks for the help!
--
Juha Pajula,
Research Assistant,
Methods and Models for Biological Signals and Images group of Signal
Processing department in Tampere University of Technology,
Finland
Ivana Drobnjak wrote:
> Hello,
> You should compare the values in the 4D activation file (pertrbation in T2*) to the T2* values in the MRpar matrix ( there is one for each tissue type and they are in column 2. Check this on the possum website, MRpar matrix structure is under inputs. ). Your 4Dactivation file values should not have a larger magnitude than the values in the MRpar matrix . Also , make sure you are using the patched version of POSSUM . You can see whether it is available on FSL download website.
> Best ,
> Ivana
>
>
Mark Jenkinson wrote:
>Hi,
>
> I can't really answer the question of whether the values in
> your maps are too large unless I know what the values in
> your maps actually are! Can you tell us what values you
> have inserted?
>
>All the best,
> Mark
> On 1 Jul 2010, at 08:59, Juha Pajula <[log in to unmask]> wrote:
>
>> Hi!
>>
>> Could you help me with the 4D activation in POSSUM fMRI simulator?
>>
>> I got following warning in node logs with every simulation I have tried:
>>
>> Creating activation4D volume and timecourse vector...
>> 4D activation mode
>> Testing if the values in the activation volumes are smaller than the T2* values
>> vol= 0
>> WARNING: Perturbation in the T2* values bigger than the T2* values.
>>
>> And after this all processes are terminating.
>>
>> I have our own data in use so probably there is some error in my input data; maybe in 4D activation T2* maps? What the error really means, are the values in the maps too large?
>>
|