Dear Nicolo,
It looks like you are trying to apply this mask to the lower-level analyses.
If you've made the mask in standard space then you should be applying it
in the higher-level analysis. It doesn't matter if the lower-level results are
masked or not, since the higher-level analysis uses the unthresholded,
unmasked statistics as inputs regardless of what is specified in the lower-level
post-stats tab.
One other thing - you should not be drawing your masks based on the
statistical analysis of the same data! This is circular and invalidates the
statistics. If you have an anatomical hypothesis for the ROI, then draw
your mask based on the anatomy, not based on the statistics.
All the best,
Mark
On 20 Aug 2011, at 14:14, Nicoḷ Cardobi wrote:
> Hi at all,
>
> I'm conducting a fMRI study on 6 kids who underwent monolateral brain ischemic lesion.
>
> I'm particularly interested in the primary motor cortex and supplementary motor cortex, so i thought about using a mask for those regions.
>
> I tried to draw a mask in fslview based either on the standard atlas or based on tresh_zstat calculated at higher (unmasked) level analysis but there is one problem as I see in the log file:
>
> WARNING:: Inconsistent orientations for individual images in pipeline!
> Will use voxel-based orientation which is probably incorrect - *PLEASE CHECK*!
>
> Image Exception : #3 :: Attempted to multiply images/ROIs of different sizes
> terminate called after throwing an instance of 'RBD_COMMON::BaseException'
> Aborted
>
> So the results of the masked analysis is exactly the same of unmasked ones.
>
>
> I already check in the archives, but the my situation seems slightly different.
>
> The FEAT input images were processed and registered to MNI 2mm brain template.
>
> The mask i draw has this values:
>
> nico@nico-EX58-UD5:~/Desktop/Mask$ fslinfo test.nii.gz
> data_type INT16
> dim1 91
> dim2 109
> dim3 91
> dim4 1
> datatype 4
> pixdim1 2.0000000000
> pixdim2 2.0000000000
> pixdim3 2.0000000000
> pixdim4 0.0000000000
> cal_max 4.0000
> cal_min 0.0000
> file_type NIFTI-1+
>
>
> While the tresh_zstat are:
>
> nico@nico-EX58-UD5:~/Desktop/Mask$ fslinfo thresh_zstat4.nii.gz
> data_type FLOAT32
> dim1 91
> dim2 109
> dim3 91
> dim4 1
> datatype 16
> pixdim1 2.0000000000
> pixdim2 2.0000000000
> pixdim3 2.0000000000
> pixdim4 1.0000000000
> cal_max 0.0000
> cal_min 0.0000
> file_type NIFTI-1+
>
> Instead, the filtered_func_data.nii.gz out FEAT prestats has those values:
>
> nico@nico-EX58-UD5:~/Desktop/Mask$ fslinfo z.nii.gz
> data_type FLOAT32
> dim1 64
> dim2 64
> dim3 36
> dim4 100
> datatype 16
> pixdim1 3.5156250000
> pixdim2 3.5156250000
> pixdim3 3.7499618530
> pixdim4 3.5000009537
> cal_max 0.0000
> cal_min 0.0000
> file_type NIFTI-1+
>
> The question is... this error occur because I create I mask on standard 2mm space instead on the (different in dimension) filtered_func_data, or there are other problems?
>
> Thanks a lot.
>
> P.s.: my fsl version is 4.1.8 on Ubuntu system.
>
|