Hi,
I came across this paper: http://www.neuro.uni-jena.de/pdf-files/Altaye-NI08.pdf where data is normalized to standard space - image dimensions of 157x189x156. But somehow I could not find this template in any toolbox, I checked SPM, FSL, MRIcro, Cat2?
Any help would be really appreciated.
They created their own templates. Refer to section "Processing: template and a priori data construction" (see right column on page 722) and Figure 1 for more details.
Hi Sahil
It may be more appropriate if you can explain to us exactly what you are trying to achieve. Image dimension, orientation, and normalization is a complicated topic. If you are simply creating your own template for normalizing a set of brains, the dimensions of the template can be set and new datasets aligned and resampled to the same space as the template. If you are trying to maintain a relationship with an established atlas, such as MNI, this is a bit more complicated to change the template dimensions while preserving coordinate space. If you can be more specific about your goal, more guidance can be provided.
Erik
Hi Chris and Dr. Middlebrooks,
Thanks a lot for your replies. Here I wanted to overlay few ROIs defined in MNI space of 157x189x156 dimensions on to same dimension MNI space. These ROIs were not clear to me how and why these dimensions were chosen so I was looking for a template of same dimensions. After downloading and checking several softwares, I figured out that the closest dimensions are of Talairach Colin27. NeuroImage 17:922-927, 2002 (161x191x151), so I just resliced this template for my purpose.
Thanks again for your time.
I may be incorrect in understanding, but it sounds like you have brain(s) with select ROIs that you want to convert the ROIs to normalized MNI space on an MNI template? If that is the case, even if your images are in an unusual dimension, you can register your brain to the MNI template (such as FSL FLIRT) and thereby convert your images to the same dimension as any MNI template. You can then apply the resultant transform matrix to your ROI files to transform them to the same dimension as the MNI template. Reslicing a template may cause you some problems if you want the coordinates to remain in true MNI space so I would be cautious with that approach. Also, you should realize there are differences between MNI and Talairach so be sure you are no interchanging the two.
If I am incorrect in assuming what you are trying to do, please correct me and we can try to help you further. Also, if you need specific help with the commands to do this, don't hesitate to ask and we can help you. I am more savvy with FSL, but if you are set on SPM, there are certainly people here who can help you with that.
Erik
Dear Dr. Middlebrooks,
Yes, you are absolutely correct. I co-registered and resliced Talairach Colin27 template with standard template (91x109x91) from FSL- made sure both are well co-registered. I further used the transformation matrix generated from this co-registration to my ROIs which were in Talairach Colin27 space to get those in standard (91x109x91) space.
Dr. Middlebrooks, I do have a question about FSL, would you mind sharing your email address in my inbox?
Thanks.
Great! Glad it worked for you. The world of orientations, dimensions, and standard space can be tricky and easy to make mistakes!
- Badges
- Science topic
- Similar topics
- Correction