 |
 |
H Jeff - I think this is fine. Cheers T On 28 Jan 2008, at 20:36, Jeffrey Spielberg wrote: > Hi, great, thanks. I have another question about Fsl-vbm/Randomise > usage. Are there some designs that are appropriate for use in Feat > but not in Randomise (in Fsl-vbm in particular)? Specifically, we > want to use a one group design correlating two questionnaire scores > with grey-matter. We set this up similarly to the example on the > Feat webpage, for example, > > Group Mean Quest_1 Quest_2 > 1 1 2.6 9.4 > 1 1 7.3 5.7 > 1 1 4.9 4.3 > 1 1 4.6 3.6 > 1 1 9.2 8.6 > > with one cope for the mean, one for each questionnaire, and one for > the difference between the two questionnaires. Is this design > appropriate for Randomise/Fsl-vbm? Thanks, > Jeff > > On Jan 26, 2008 1:21 AM, Steve Smith <[log in to unmask]> wrote: > Hi, > > On 24 Jan 2008, at 21:57, Jeff Spielberg wrote: > > > Hi, I have a couple of questions regarding Fsl-vbm and First. The > > scans we > > have been using for both have some signal loss superiorly and > > inferiorly. > > Due to this we do not get a good extraction with Bet (brain is > > excluded top > > and bottom). We have been dealing with this issue by correcting for > > the > > bias field with Fast and have gotten good results. Since Fsl-vbm > > applies > > Bet we have entered bias-corrected scans into our Vbm analyses. > > However, > > will this cause problems when segmentation is done in > > fslvbm_2_template (or > > anywhere else in Fsl-vbm)? > > The bias correction is fine, but if you have already run BET that > might compromise the FSL-VBM when you feed the already-betted image > into FSL-VBM. The simple solution is to apply the bias field > correction to the original image and feed that into FSL-VBM. That > should be fine then. > > > Also, is it best to use the bias-corrected scans in First as well? > > It seems > > possible that not correcting scan intensities for the bias field > will > > interfere with the identification of structures (since First uses > > intensities in this process), thanks, > > If it looks like the bias field correction has improved the image then > it should be better to feed the corrected image into FIRST than the > original. > > Cheers, Steve. > > > > > > Jeff > > > > > ---------------------------------------------------------------------- > ----- > Stephen M. Smith, Professor of Biomedical Engineering > Associate Director, Oxford University FMRIB Centre > > FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK > +44 (0) 1865 222726 (fax 222717) > [log in to unmask] http://www.fmrib.ox.ac.uk/~steve > ---------------------------------------------------------------------- > ----- >