[FieldTrip] failed segmentation

Lilla.Magyari at mpi.nl Lilla.Magyari at mpi.nl
Sun Nov 17 12:03:15 CET 2013


hi Vitoria,

is this a tpm (the default output of ft_volumesegment)? Have you also
tried to segment this mri with an output for 'brain','skull','scalp'?
because when you have a binary representation for these tissue-types, they
do not overlap, and are closed. So, I am wondering if you end up with a
brainmask which do not contain the artifact at the top of the head.


Lilla


> Hi Vitoria,
> it looks like there is an artifact on the top of the head. It happened to
> me already and I had to preprocess the mri volume by using an appropriate
> cutoff and by thresholding (after plotting the histogram of the grey
> levels).
>
> mri.anatomy = (mri.anatomy>thr).*mri.anatomy;
>
> or
>
> mri.anatomy(:,:,230:256) = 0;
>
> If thresholding is still not effective you might consider applying
> morphological operators (erode, dilate, open, close), after taking care
> that you voxel dimensions are cubic (diagonal values of the canonic form
> of
> mri.transform(1:3,1:3) are all the same). I use ft_volumereslice for that.
>
> I hope this helps,
> Cristiano
>
>
>
> On Sat, Nov 16, 2013 at 9:31 AM, Vitoria Piai
> <v.piai.research at gmail.com>wrote:
>
>> Hi all,
>>
>> has anyone ever had a failed segmentation that looks like the one in the
>> figure?
>> My script works for all participants, I mean, for all other participants
>> the brain is only inside the skull :)
>> But it fails terribly for this one. I thought it could have to do with
>> the
>> placement of the fiducials, and I've played around a bit with that but
>> it
>> didn't solve the problem.
>> If anyone recognises this type of error, I'd be help to hear if/how it
>> can
>> be fixed.
>>
>> Thanks a lot, Vitória
>>
>> _______________________________________________
>> fieldtrip mailing list
>> fieldtrip at donders.ru.nl
>> http://mailman.science.ru.nl/mailman/listinfo/fieldtrip
>>
> _______________________________________________
> fieldtrip mailing list
> fieldtrip at donders.ru.nl
> http://mailman.science.ru.nl/mailman/listinfo/fieldtrip





More information about the fieldtrip mailing list