[Mristudio-users] Questions about Diffeomap and ROIeditor.

[susumu mori]

Please see below;



> 1) I have anatomical images and DTI images, all in DICOM format. And I want
> to know how to superimpose these two types of images (crop and rescale)?.
> Here is what I did.I tried to use Diffeomap, I use the "Crop" button to I
> throw away some axial slices from the anatomical images. Then I use
> "Resample Image" button, and I enter the dimensions of the DTI image and I
> get a part of the image (the scaling step was not right? It crops the right
> and left sides of the image)
>

> Crop tool is to prepare two images before submission to non-linear
transformation, but not to do resampling. For example, if your image covers
only to the level of the midbrain and out atlas is extended to the medulla.
Then, non-linear image-matching algorithm (LDDMM) gets confused. In this
case, you want to crop the atlas image to the level of the midbrain so that
both images covers the same areas of the brains.
> You can match the pixel size and matrix size using the "resample image"
tool. However, this works only when; 1) the center of FOV is identical
(translation) and 2) the orientation of the FOV is identical (rotation). If
T1 and DTI were acquired without changing the center of FOV and it's
orientation (and also if the subject motion is negligible), you can
co-register two images just by the resample tool
> If your T1 and DTI were not co-registered when they were acquired, you
need post-processing registration. There are 1) theoretical registration and
2) registration tool (AIR)
> For theoretical registration, you need to extract information about the
center of FOV and oblique (rotation) information from DICOM or other types
of header information. Then you can type in these information in the
"translation" and "rotation" tools in DiffeoMap.
> For registration tool, you need to use "rigid transformation" of AIR. Use
DTI (such as b0 or DWI) as a template and register T1 to the template. Here,
you can submit T1 image even if the matrix, pixel size, brain positions are
different from DTI. The output file has the same dimensions as DTI. This
procedure is explained in
https://www.mristudio.org/wiki/user_manual/diffeomap/GetStarted_StepOne_LinearNormalization.
As you can see, this could be the easiest way, but please make sure that the
output of this automated registration actually worked by carefully inspect
the results. If the initial position of the brain in T1 is very different
from DTI, it may not work. You many need to use "mutual" registration (it
takes much longer) or few-point landmarks (manual but much quicker) for
initial alignment or sometimes you need to remove skull intensity. All these
operations can be done within DiffeoMap.


> 2) When I tried to define some ROIs in the ROIeditor and save my ROI data
> as "binary maps", then try to open this map in DTI Studio, I get a message
> that tells me *"Unrecognized image-ROI data file!"*. I don't know what
> this means.
>

> please use "binary" not "binary map". To learn what is the "binary map"
format, please refer to FAQ#16. You need to save an ROI as "binary" in
RoiEditor and open is as "binary" in DtiStudio. If you have only 1 ROI, this
is the best way. If you have multiple ROIs and you want to combine them with
specific operations such as "OR', "AND", "NOT", then you want to use "binary
map". You need to save the ROI files as "binary map" in RoiEditor and load
as "binary map".


> 3) Also, can I define the ROIs and the seeds in the anatomical image then
> transform it to the DTI images?
>

> Yes, first T1 has to go through DiffeoMap to make sure it is coregistered
to DTI and has exactly the same matrix and pixel dimensions (AIR rigid
registration could be the easiest way, see above).
> Then load the registered T1 to RoiEditor and define ROIs. If you define
one ROI, save and load as "binary". If you draw multiple ROIs and combine
them for tractography, use "binary map".


> By the way, I can not use the brain Atlas that comes with the software,
> because I'm not dealing with brain data. Thank you.
>
> Fatma.
>
>
> _______________________________________________
> Mristudio-users mailing list
> Mristudio-users at mristudio.org
> http://lists.mristudio.org/mailman/listinfo/
> Unsubscribe, send a blank email to:
> Mristudio-users-unsubscribe at mristudio.org
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.mristudio.org/pipermail/mristudio-users/attachments/20100617/c5592904/attachment.html