Hi,
You want to be using the PVE values and summing up the
partial contributions. So what you want is the sum of the
PVE values across the image, including those with PVE
values less than 50%, as these contributions, when summed
up, can be significant. The commands you are using only do
voxel counts which isn't nearly as accurate.
The relevant lines from the sienax script (in FSL4.0) are:
S=`${FSLDIR}/bin/fslstats ${I}_stdmaskbrain_pve_1 -m -v`
xa=`echo $S | awk '{print $1}'`
xb=`echo $S | awk '{print $3}'`
ugrey=`echo "2 k $xa $xb * 1 / p" | dc -`
ngrey=`echo "2 k $xa $xb * $vscale * 1 / p" | dc -`
echo "GREY $ngrey $ugrey" >> report.sienax
This takes the stdmaskbrain_pve_1 (or 2 later in the script)
and calculates the mean pve value over the whole image
then multiplies this by the whole image volume, giving the
volume of this tissue. The variable vscale is equal to the
scaling factor of the affine transform to standard space,
and so converts the volumes into ones normalised to the
standard space template (ngrey is normalised, ugrey is not).
I hope this helps.
All the best,
Mark
On 30 Dec 2007, at 04:45, Bettyann Chodkowski wrote:
> Hello All,
>
> I am trying to recalculate the 'brain percentage' value calculated
> by sienax as reported in the
> *.sienax file using the volumes created by sienax.
>
> for example, i have these values in my *.sienax file:
>
> Class: CSF tissue 1 tissue 2
> brain percentage
> Volumes: 316748.7 516990.9 384503.5
> 0.739996
>
> i was hoping to get the same volumes using avwstats++ on the PVE
> volumes but do not:
>
> % avwstats++ 20071129_13_1_stdmaskbrain_pve_0 -l 0.50 -V
> 370841 297463.903287
>
> % avwstats++ 20071129_13_1_stdmaskbrain_pve_1 -l 0.50 -V
> 665191 533571.830762
>
> % avwstats++ 20071129_13_1_stdmaskbrain_pve_2 -l 0.50 -V
> 482694 387184.915728
>
> i tried using the 'seg' volume but do not:
>
> % avwstats++ 20071129_13_1_stdmaskbrain_seg -l 0.99 -u 1.01 -V
> 428073 343371.594489
>
> % avwstats++ 20071129_13_1_stdmaskbrain_seg -l 1.99 -u 2.01 -V
> 641821 514825.976286
>
> % avwstats++ 20071129_13_1_stdmaskbrain_seg -l 2.99 -u 3.01 -V
> 448832 360023.079002
>
> How might I go about recalculating the 'brain percentage' value
> reported by sienax?
>
> Happy New Year,
> - BettyAnn
>
|