Hi Mark,
thanks a lot for your reply. I need to clarify one more thing:
when doing the BET manually, I always put the coordinates for centre
of initial brain surface sphere
that I set up for that particular individual in fslview. Do you think
adding something like "find centre of gravity for each particular
brain" when running the BET script would help? Do you happen to have
some experience with that?
Thanks a lot,
Klara
On Tue, Aug 31, 2010 at 3:36 AM, Mark Jenkinson <[log in to unmask]> wrote:
> Hi,
>
> Your bet call looks fine.
> If you are not using any options in the GUI then
> you should get exactly the same output using
> the GUI or this script. If that is not the case then
> something is very odd and you should carefully
> check that you are not using any different options
> in the GUI and also check that the version of FSL
> that your script is calling is the same as you are
> using for the GUI (e.g. use "which bet" to find out
> what your script would be calling).
>
> It is not unusual to have to tailor the bet options
> to certain individuals. You might want to try the
> -R option in bet as that sometimes does a
> better job over many datasets. However, if you
> do need to tune your parameters for individual
> cases, then you will need to do this by hand
> and not via a script anyway, as the only good
> way of telling whether it works well is to look at
> the results manually.
>
> If you are still having trouble or continue to get
> differences between your script and GUI runs
> then let us know.
> All the best,
> Mark
>
>
> On 30 Aug 2010, at 22:32, Klara Mareckova wrote:
>
>> Hello,
>>
>> I'm familiar with the FSL GUI interface for BET and FEAT analysis of
>> the data but since we're working with very large datasets, I'm trying
>> to learn how to use scripting to do these in FSL. While the FEAT
>> analysis seems to be running smoothly as a script based on the .fsf
>> design, my BET sript has most likely a problem because about one third
>> of the subjects fail the registration! (mostly because of brain cut
>> off in the frontal or occipital lobe) This does not happen when
>> working on the same dataset manually.
>>
>> Could anybody please explain how to run BET as a script and how to do
>> the tuning so that it's able to incorporate individual differences in
>> the brain?
>>
>> Here's the BET sript I'm using now for reference:
>>
>> bet input.nii output.nii -f 0.5 -g 0
>> where
>> -f <f> fractional intensity threshold (0->1); default=0.5;
>> smaller values give larger brain outline estimates
>> -g <g> vertical gradient in fractional intensity threshold
>> (-1->1); default=0; positive values give larger brain outline at
>> bottom, smaller at top
>>
>>
>> Thanks a lot for any suggestions,
>>
>> Klara
>>
>
|