Hi,
You seem to have many questions about this which I think you'll find are answered in the following papers, the FSL Course material and the FSL wiki. I recommend that you look at this material as it can explain more than we can write in a few emails.
Papers:
• On the validation of the FSL-VBM approach and a clinical application:
see Douaud et al., Anatomically related grey and white matter abnormalities in adolescent-onset schizophrenia. Brain 2007.
• On the VBM protocol:
see Ashburner et al., Voxel-based morphometry-the methods. Neuroimage 2000.
see Good et al., A voxel-based morphometric study of ageing in 465 normal adult human brains. Neuroimage 2001.
• On the issues of such voxel-wise analyses:
see Bookstein,"Voxel-based morphometry" should not be used with imperfectly registered images. Neuroimage 2001.
see Ashburner et al., Why voxel-based morphometry should be used. Neuroimage 2001.
see Crum et al., Zen and the art of medical image registration: correspondence, homology, and quality. Neuroimage 2003.
see Jones et al., The effect of filter size on VBM analyses of DT-MRI data. Neuroimage 2005.
All the best,
Mark
On 28 Jun 2014, at 10:05, SUBSCRIBE FSL Xinfa_Shi <[log in to unmask]> wrote:
> Hi,
>
> I was following the user guide provided on FSL-VBM page , and found some sentences in fslvbm_3_proc : The script runs randomise with inference (generation of p-value maps) turned off, so that it very quickly creates just the raw tstat maps. These tstats maps should help you decide which smoothing is the most relevant to feed into a full run of randomise, and which threshold to use for the cluster-based thresholding (option -c in the randomise command) .
>
> Question :
> 1) How do tstats maps help us decide which smoothing is the most relevant to feed into a full run of randomise ?
> 2) How to select the size of smoothing kernel ?
>
> In general ,I use sigma = 3mm (it is appropriate ? is there theory basis ? ) because I run randomise using sigma=2,3,4,5,6,and find the range of activation using sigma=3mm is largest (running command "fslstats fslvbm_tfce_corrp_tstat1.nii.gz -R"). Is it right ?
>
> Regards .
>
> Shi Xin-Fa .
|