Hi - have a look in the *avscale and *sienax files output by SIENAX -
you'll see that the "VSCALING" used by SIENAX is what avscale calls
the "Determinant" - which is the correct volumetric scaling factor.
Cheers, Steve.
On 8 Jun 2009, at 18:13, Daniel Schwartz wrote:
>
> Hello all,
>
> We've recently run a number of subjects in sienax and have some
> questions about how the scaling factor for each subject is derived.
> It is our understanding that the scaling factor is the AVERAGE of
> the three scalars representing the affine transformation scaling
> parameters of the single subject's skull to the MNI skull. The
> scaling matrix looks like THIS (we think):
>
> S1 0 0 0
> 0 S2 0 0
> 0 0 S3 0
> 0 0 0 1
>
> where S1, S2 and S3 represent scaling in the x, y and z dimensions,
> respectively. Then sienax takes the average of these three numbers,
> and the "corrected" volumes that sienax spits out are the product of
> the total volume (GRAY, PGRAY, BRAIN, and CSF) and this averaged
> number. As we understand it, though, a "true" measure to describe
> the scaling of this volume to another volume would actually be the
> product of S1, S2 and S3, not the average. For instance:
>
> VOLUME A: 5 mm x 5 mm x 5 mm = 125 mm^3.
> VOLUME B: 2.5 mm x 10 mm x 20 mm = 500^3.
>
> SCALING FACTORS: S1 = 0.5, S2 = 2, S3 = 4. (S1 = B1/A1, S2 = B2/A2,
> S3 = B3/A3).
>
> AVERAGE OF SCALING FACTORS: (0.5 + 2 + 4)/3 = 2.167
> PRODUCT OF SCALING FACTORS: 0.5 * 2 * 4 = 4!
>
> ACTUAL VOLUME SCALING FACTOR: 500 mm^3/125 mm^3= 4!
>
> I guess my point is, the average scaling factor is not a real volume
> scaling factor at all, and becomes more and more "incorrect" as the
> differences between S1, S2 and S3 become larger (as in my extreme
> example above... if VOLUME A is taken to be the MNI skull, VOLUME B
> is a misshapen head indeed.)
>
> Since you folks wrote the program, we're betting we're actually
> missing something. But maybe we're not?
>
>
> Thank you again!
> -Daniel
>
---------------------------------------------------------------------------
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
---------------------------------------------------------------------------
|