Dear Shal Hat,
This registration has clearly gone very, very wrong.
The solution is to fix the registration, not to try anything like
fslswapdim.
The output images should look like each other and be in the same
alignment, regardless of the labels. You should check the labels on
the original images (prior to registration) to make sure that they are
correct and fix the labeling then, if necessary. It should not be fixed
after registration.
As for determining left/right without a marker (e.g. vit E capsule) - in
general you cannot. This is why it is so important to make sure that
the original images are correct (preferably by checking with a marker).
Once you have a method for reconstructing the images which works
correctly for the labels, then always use this (for that study/protocol)
and everything should be fine.
To fix your registration may require some extra effort after the
labels are fixed. The images you sent look strange. Are the
original images standard brain images?
All the best,
Mark
On 29 Nov 2009, at 16:13, Shal Hat wrote:
> Actually when displaying the acquired original 3D anatomical hdr
> file in fslview
>
> In the top left window I have the axial slices labeled with the
> anterior part labeled R, the posterior labeled L, the R/L is labeled
> S/I.
> In the top right window I have a coronal view where superior is
> labeled posterior, the inferior is labeled anterior, and the R/L is
> labeled S/I.
> In the lower left window I have sagittal view where the anterior is
> labeled right, the posterior is labeled left, the inferior is
> labeled anterior,the
> superior is labeled posterior.
>
> How can I solve this using fslwarpdim or other tools ?
>
> Also, how may I know which part is right/left, with/without "vitamin
> E" capsule?
>
> Would I have to do any reorientation for the functional images too?
> If I reorient the 3D anatomical?
>
> Given the above orientations would I be able to leave it as it is
> and there will be no mismatch between the labeling
> and orientation in the first and higher level analysis?
>
> Thanks a million :)
>
>
> On Wed, Nov 11, 2009 at 10:40 AM, Matthew Webster <[log in to unmask]
> > wrote:
> Hi,
> Is the brain mislabelled ( e.g. do the AP LR labels match how the
> brain is displayed )? If not then you can use fslswapdim to
> reorganise the storage order of the image.
>
> Many Regards
>
> Matthew
>
>
> Hi,
>
> I use BET to extract the brain. However when I looked at it using
> FSLVIEW, the brain was
> inverted (cortex on bottom, anterior is where Right should be and
> posterior is where Left
> should be). How would I be able to resolve this issue? What caused
> it? How will it affect my
> GLM or ICA analysis?
>
> Thanks a million :)
>
>
|