It looks like it fit the GLM through the data, but simply didn't find
anything significant. I would suggest checking the mask.img to see what
was included in the analysis, and the t statistic, contrast and ResSS
images to look for clues about why nothing seemed significant.
Best regards,
-John
On Thu, 2009-06-25 at 16:55 +0100, Benetti, Stefania wrote:
> Dear SPMers and John
>
> Further the previous posts about using DARTEL for longitudinal VBM,
> we
> decided to follow the HDW approach that Kipps used. Briefly:
>
> 1. co-registration and HDW of late (FU) to early images (BA) for each
> subject.
> 2. save Jacobian determinats (JY)
> 3. Segmentation of BA images
> 4. BA used to create a DARTEL template
> 5. BA images and JY warped with template ( modulation=yes)
> 6. wmBA multiplied by wmJY-> product images ( this using ImCalc)
> 7. normalisation to MNI
> 8. smoothing
>
> So far everything run smoothly.
> However, after having specified the design, I could not estimate the
> model. SPM returned this error message" check your data. There are no
> significant voxels." A plot of the globals shows values below 0.4
> I tried to lower the masking threshold and to set it to 'none' but
> SPM
> could not estimate the model either.
>
> The same happened earlier when just using FU-BA difference images
> obtained from DARTEL pre-processing, though I was able to estimate
> the
> model after setting the masking threshold to 'none'
> It seems to happen every time I use ImCalc on images normalised to
> DARTEL template.
> Has anyone encountered this before?
>
> Many thanks
> Stefania
>
> **********
> Stefania Benetti MSc
> PhD student
> Section of Neuroimaging, Box P067
> Institute of Psychiatry - King's College London
>
>
>
> The suggestion I made was about how to use DARTEL to do the intra-
> subject
> alignment, so I was basically just answering the question. I have no
> empirical evidence either way, but (for the reasons I mentioned) I
> suspect
> that HDW may do a marginally better job with most serial scans.
>
> Best regards,
> -John
>
> On Friday 22 May 2009 11:15, Benetti, Stefania wrote:
> Thank you John for your answers and suggestions. However, I am not
> sure to
> understand what you are suggesting about HDW approach that Kipps
> used.
>
> "In general, I think I would still suggest the HDW approach that
> Kipps and
> others have used...For inter-subject alignment, the residuals are
> rarely
> i.i.d. Gaussian, which is why I chose to align tissue class data
> instead.... However, the segmentation errors may be relatively large
> compared with the volumetric differences between the serial scans,
> which
> would make the DARTEL approach less accurate"
>
> My understanding was that the procedure you suggested to Reinders
> (https://www.jiscmail.ac.uk/cgi-bin/wa.exe?A2=ind0804&L=SPM&P=R48484)
> consisted in an adaptation of Kipps' approach for DARTEL, in which
> the
> within subject alignment approach replaced the HDW. Am I completely
> wrong?
> Did you mean that this longitudinal DARTEL procedure may be less
> accurate
> than the procedure that Kipps actually used?
>
> Many thanks again
> Stefania
>
>
>
>
> -----------------------------
>
> Stefania Benetti BSc MSc
> Research assistant
> Institute of Psychiatry
> Neuroimaging Section, Box P067
> De Crespigny Park SE5 8AF
> London
>
> tel: +44 (0)20 7848 0284
> fax: +44 (0)20 7848 0976
> -----------------------------
>
>
>
--
John Ashburner <[log in to unmask]>
|