Hi,
>
> 1.I translated the data from .par .rec in compressed FSL 4D Nifti
> Nii by ,like
> written in http://www.fmrib.ox.ac.uk/fslfaq/#general_3d4d, Mri Cron
> DICOM to
> Nifti Converter (http://www.sph.sc.edu/comd/rorden/mricron/dcm2nii.html
> )
> (the option reorient large images to nearest ortoghonal plane is ON)
> I can put these resulted nii.gz in the FEAT analysis or I have to
> change the
> left/right orientation (I readed about neurological-radiological
> convention..)?
>
> 2.In FEAT, Stats, Full model setup,I hev done 2 EVs, with Basic
> Shape=Square
> (is OK?).EVs A (the first to start) is with Skip=0, On=60 sec,
> OFF=60 sec, the
> EVs B (that starts after the A) is with Skip=60 seconds,On=60 sec,
> OFF=60
> sec, is OK?
>
In general, take a look at the resulting EV when you click on View
design on the GLM setup GUI to see if it looks sensible to you.
In this case it looks like you have just 2 conditions A and B, and you
are modelling both separately with 2 different EVs. This is not the
right thing to do. There is only ever information in BOLD-FMRI in the
signal _changes_ between different conditions. This means that there
is no information in the mean of the FMRI time series (hence why we de-
mean the data and design as a preprocessing step) and so you need to
leave the mean signal unmodelled. In your case this translates into
just having _one_ EV, which is "on" for condition A and "off" for
condition B (or vice versa). This is what you should have seen when
you used the Model Setup Wizard with rArA (which in your case
corresponds to ABAB), i.e. only 1 EV is modelled (along with its
temporal derivative).
> 3.I have to change also the Phase? (or is not central?) Stop after=
> -1,
> Convolution=Gamma, is OK?
>
Setting the phase will shift the square wave in time (take a look at
the bubble help) - so it depends on your experimental timings.
Gamma convolution should be fine.
> 4.In the Model Setup Wizard rArArA..is OK? I have done rest
> period=60 sec, A
> period 60 sec, is ok?
>
yep.
> 5. In data high pass filter cutoff=130 (more than 60+60=120),is OK?
yep - that is fine.
Cheers, Mark.
----,
Dr Mark Woolrich
EPSRC Advanced Research Fellow University Research Lecturer
Oxford University Centre for Functional MRI of the Brain (FMRIB),
John Radcliffe Hospital, Headington, Oxford OX3 9DU, UK.
Tel: (+44)1865-222782 Homepage: http://www.fmrib.ox.ac.uk/~woolrich
On 6 May 2009, at 17:00, SUBSCRIBE FSL Alex rigon wrote:
> Hi I am doing a single run of 12 minutes, composed by 2 conditions,
> condition
> A of duration of 60 seconds, condition B duration of 60 seconds. The
> condition
> A will start first (so the sequence will be ABABABABABAB ). I need
> to know:
>
> 1.I translated the data from .par .rec in compressed FSL 4D Nifti
> Nii by ,like
> written in http://www.fmrib.ox.ac.uk/fslfaq/#general_3d4d, Mri Cron
> DICOM to
> Nifti Converter (http://www.sph.sc.edu/comd/rorden/mricron/dcm2nii.html
> )
> (the option reorient large images to nearest ortoghonal plane is ON)
> I can put these resulted nii.gz in the FEAT analysis or I have to
> change the
> left/right orientation (I readed about neurological-radiological
> convention..)?
>
> 2.In FEAT, Stats, Full model setup,I hev done 2 EVs, with Basic
> Shape=Square
> (is OK?).EVs A (the first to start) is with Skip=0, On=60 sec,
> OFF=60 sec, the
> EVs B (that starts after the A) is with Skip=60 seconds,On=60 sec,
> OFF=60
> sec, is OK?
>
> 3.I have to change also the Phase? (or is not central?) Stop after=
> -1,
> Convolution=Gamma, is OK?
>
> 4.In the Model Setup Wizard rArArA..is OK? I have done rest
> period=60 sec, A
> period 60 sec, is ok?
>
> 5. In data high pass filter cutoff=130 (more than 60+60=120),is OK?
>
|