[Brains-users] where to cut the brain stem

[Ronald Pierson]

My opinion is that it probably doesn't make a big difference, unless you
have a real interest in brainstem CSF.  To my knowledge, we haven't ever
used brainstem volumes directly from these measures for anything.
Indirectly, one could use them to remove the brainstem and cerebellum
from whole brain volumes to get cerebral volumes.  For that purpose, the
definition of the cutoff is irrelevant - it all gets subtracted.  Also,
adding up the cortical lobes does the same thing, and you don't need the
brainstem for that, either.  Maybe it will have a small affect on total
brain compartment volume, if you use that for anything.

However, that doesn't really help you make a decision about the cutoff.

Vince Magnotta suggested something a few years ago that makes alot of
sense.  Why not cut the brainstem at the bottom of the extended
talairach bounds?  We add two rows of boxes below the normal talairach
boxes to include the cerebellum.  To see them, load the Talairach.bnd,
go to the selector, select the talairach bounds, click on object
properties, then turn on the grid.  Then, simply trim the brain trace to
cut off anything of the spinal cord/brainstem that extends below this.
If the trace doesn't extend down far enough to reach the bottom of the
grid, you will need to extend it down there.  I think this does a nice
job as a guide and really takes away the hassle of finding the arteries,
figuring out what to do when they don't look normal, etc.

We haven't implemented it here because we have lots of longitudinal
scans and need to continue doing it the same way to maintain continuity.
However, if I were starting from scratch on a dataset, that is how I
would do it.

Ron

On Tue, 2006-03-21 at 16:37 +0100, Christoph Christmann wrote:
> Hi all,
> 
> taking into account the manuals procedure to cut the brain at the 
> vertebral arteries I end up with a brain stem cutted oblique and the 
> lower part of the cerebellomedullar cinsterna is missing. However, if 
> I add the complete cisterna the measured occipital CSF is clearly 
> bigger than the Iowa norm. I am very interested in the procedures of 
> other groups - it seems necessary to me to include the whole cisterna 
> volume to get an overview over the CSF. But regarding the inclusion 
> of parts of the brain stem I am not quite sure.
> 
> TIA
> 
> Christoph
> 
> 
> 
> 
-- 
Ronald Pierson <ronald at psychiatry.uiowa.edu>