Hi - yes you can expect to see significant atrophy in both groups and
strong correlation between SIENA and SIENAX.
I'm guessing that your data is maybe problematic, particularly as 5
years is a long time in the life of an MRI machine, and the two
timepoints may be hard to combine unless your physicists have been
very careful to keep things stable.
If you'd like to upload an example image pair with SIENA output we
can take a look.
Please upload the files in a single compressed tarfile to
http://www.fmrib.ox.ac.uk/cgi-bin/upload.cgi
and then email me the upload ID.
Cheers
On 30 Aug 2007, at 02:58, Antonios - Constantine Thanellas wrote:
> Dear fsl users, I used sienax, siena and voxelwise siena for cross
> sectional, longitudinal and statistic analysis between two groups
> Controls
> and Parkinsonians.
>
> From the 8 sets (each set of 2 scans at intervals of 0 and 5 years) of
> Controls and after the use of sienax 6 of them showed growth in NBV
> ( the
> 5th year’s sienax result had bigger NBV than the 0th year's sienax
> result)
> and only 2 of them showed atrophy. I was expecting to see atrophy
> in all of
> them so I checked if I did a mistake by considering the scans of
> the 5th
> year as the scan of the 0 year but everything was correct. Normally
> the
> scans of 5th year should show less NBV than the 0st year’s scans, Am I
> right? What’s your opinion?Why do I have results that shows growth
> instead
> of atrophy?
> At the same time the results of siena in all Controls showed growth
> rates
> instead of atrophy!!Why is that? (I used the set up: siena “0 year
> scan”
> “5th year scan” . With the set up siena “5th year scan” “0 year
> scan” I had
> almost the same PBVC as previously but in this case showing
> atrophy. Which
> setup is the right one and why?)
>
>
> According to S.M Smith et al./neuroimage 36 (2007) 1200-1206 siena and
> sienax -especially in NBV and PBVC- exhibit high correlation but
> why in my
> case I have such uncorrelated results ??(in both siena set ups
> there are
> cases where PBVC and NBV give totally opposite information. For
> instance NBV
> shows growth and PBVC atrophy.. Which results should I “trust”??)
>
> As far as it concerns the Patients group from the 13 sets of data
> 12 of them
> showed growth of PBV and only 1 showed atrophy while siena ( with
> the set
> up: siena “0year scan” “5th year scan”) showed growth is 7 cases
> and atrophy
> in 6.
>
> Totally unexpected values for both Controls and Patients. Do you
> have an
> idea on why this happens?
>
> I also run voxelwise siena and found some evidence of atrophy among
> patients
> (compared to the healthy subjects) but since the results of siena –
> sienax
> are bizarre , I can’t rely on the results of randomise, can i? since
> randomise uses as input the results of siena then these significant
> areas of
> atrophy among patients are wrong…Am I right?
>
> Correct me if I’m wrong , normally I should expect NBV in Controls
> to reveal
> atrophy between older and new scans , a result that should agree
> with the
> siena results and as far as it concerns patients more atrophy (in
> both siena
> and sienax) should be expected comparing to controls..
>
> I would really appreciate your help
>
>
> Thanks a lot for your time
> Antonios-Constantine Thanellas
------------------------------------------------------------------------
---
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
------------------------------------------------------------------------
---
|