Please provide more details as far as the image size and spatail resolutions of your data (native space), the size of the Dratel template. Based on the copy I have of the standard ch2bet, the size is actually different from what Dartel adopts, but clearly the reslicing seems to be using different template.
Additional Suggestions
1. As a troubleshooting hint, why not try first the standard procedures on a small subset without Dartel and then use Dartel to compare and isolate.
2. In my implementation of DTI-based VBM (will release a special Multifaceted DTI-Toolbox in June), I do not use the T1w data at all, there are several reasons for this choice and many publications actually did this already (check previous emails). The b0 is used for both intra-registration (diffuseion eddy or geometric distortion corrections) and then as an object for the normalization with Target being the EPI template and the others are the "DTI-children maps -FA, diffusivities,..". The b0 has contrast similar to the EPI template and the registration works just fine. If this procedure is implemented carefully you will get results as good as TBSS which focusses the attention on the central white matter common between all subjects and hencce normalization issues are not as destructive. For this task both Dartel and the classical normalization were just fine, differences are mostly related to processing options.
Khader M Hasan, PhD
Associate Professor of Radiology
MSE 168, Tel 713 500 7690 (FAX 713 500 7684)
University of Texas Health Science Center at Houston
Medical School
Diagnostic and Interventional Imaging
Magnetic Resonance Imaging Research Division
Diffusion Tensor Imaging Lab, Tel 713 500 7683
http://www.uth.tmc.edu/radiology/faculty/khader-m-hasan/index.html<http://www.uth.tmc.edu/radiology/faculty/hasan.html>
________________________________
From: SPM (Statistical Parametric Mapping) [[log in to unmask]] On Behalf Of Gao Junling [[log in to unmask]]
Sent: Monday, April 09, 2012 9:43 AM
To: [log in to unmask]
Subject: [SPM] Children DTI has a bigger spmT001 result than MNI template
Dear SPM Experts,
We are doing DTI VBM analysis on Children with VS. without epilepsy. The procedures are following standard ones:
1. Individual DTI was coregistered to individual T1
2. The Individual T1 is normalized to the template in VBM using DARTEL
3. The deformation matrix created during T1 normalization is applied to normalize the individual DTI parametric images (FA, MD, etc).
4. Normalized parametric images is smoothed
5. Group comparison using 2-sample Ttest in SPM.
The processes were fine. however, when we carefully checked the results, we found the the spmT001.img file looked bigger than the standard MIN template (using ch2bet.nii) (http://trackvis.org/blog/wp-content/forum-image-uploads/gaolion/larger.png).
We checked the smoothed and normalized parametric images, and found all of them had similar size to MNI. Now we are quite confused about that the final results, why it became bigger size than MNI. We even tried to make a template using our own children data, and registered all the T1 images to this template, but still we have the same problem.
[http://trackvis.org/blog/wp-content/forum-image-uploads/gaolion/larger.png]
Anyone has any hint for the bizarre result? Many thanks in advance!
Best wishes,
Gao
|