Hi Bruce,
it seems I have uploaded the dir as well.
Thanks for your help!
Regards
Antonio
Messaggio originale
Da: cra...@inwind.it
Data: 25/10/2013 17.02
A: fis...@nmr.mgh.harvard.edu
Cc: freesurfer@nmr.mgh.harvard.edu
Ogg: [Freesurfer] R: Re: R: Re: analysis on 4-6yrs pat
Hi Bruce,
Hi all,
Apologies for this basic question, but im quite the newbie.
I'mcurrently using tksurfer and got this result:
surfer: filling (ctrl-c to cancel) done, 4646 vertices filled
surfer: performing FDR on 1989 vertices
MRISfdr2vwth(): np = 1989, nv = 163842, fdr = 0.05, vwth=2.24098
You can use mri_label2vol to create a nii mask from the label. To map
the label to the DTI space, register your DTI (b=0) to the anatomical
using bbregsiter. You can then use mri_label2vol to transfer the label.
doug
On 10/28/2013 06:14 AM, Zhao Jingjing wrote:
Hi,
Our lab have managed to
We don't have a way to do a power analysis directly from QDEC. I have
some matlab code that does it for univariate analysis.
In the QDEC output folder, you will see a glm folder. In that there will
be folders for each contrast. In the contrast folder, you will see an
F.mgh file which will be
Thanks Doug,
Would you please provide me the MATLAB code and instruction!
BR
On Tue, Oct 29, 2013 at 3:25 PM, Douglas N Greve
gr...@nmr.mgh.harvard.eduwrote:
We don't have a way to do a power analysis directly from QDEC. I have some
matlab code that does it for univariate analysis.
In the
The matlab code is here:
$FREESURFER_HOME/fsfast/toolbox/fast_glmpower.m
There's not much documentation on it. Have you done power analyses before?
doug
On 10/29/2013 10:29 AM, amirhossein manzouri wrote:
Thanks Doug,
Would you please provide me the MATLAB code and instruction!
BR
On
One comment on power calculations. You cannot do the power calculations
after you have completed the study. Any reviewer/referee asking you to
compute the power should be politely informed that post-hoc power analyses
are not valid and are biased (because you would be computing the power of
the
Hi Elisa, the fact that the surface does not follow the segmentation is
not a problem by itself. The cortical segmentation created by the
volumetric segmentation is never used. It just needs to be done as part
of the subcortical segmentation. However, tou should make sure that the
surfaces
Good point Donald. Where can one get G*Power?
doug
On 10/29/2013 10:38 AM, MCLAREN, Donald wrote:
One comment on power calculations. You cannot do the power
calculations after you have completed the study. Any reviewer/referee
asking you to compute the power should be politely informed that
I mean to take all the values (eg, .68849, .31061, etc), compute their
mean, then subtract the mean from each value. This will not affect the
slope calculation but can affect the offset. Seeing differences in the
offsets does not mean that anything is fundamentally wrong.
doug
On 10/29/2013
For anyone who wants G*Power, here is a link to the download page. It is
free for all users.
http://www.psycho.uni-duesseldorf.de/abteilungen/aap/gpower3/
Best Regards, Donald McLaren
=
D.G. McLaren, Ph.D.
Research Fellow, Department of Neurology, Massachusetts General Hospital
When I run
asegstats2table -i aseg.stats_subj2 -t junk
I get EstimatedTotalIntraCranialVol as the last column.
doug
On 10/27/2013 06:06 PM, Daniel Ferreira wrote:
Hi Douglas,
You are right, sorry for not including the thread. Below I paste our
previous conversation. I also attach again
What is the problem with small values? You can multiply them by 100 to
give you percent of brain. Or multiply them by 1000 to give you tenths
of percents
doug
On 10/24/2013 12:58 PM, Dorian P. wrote:
Thank you Douglas.
The eTIV has values like 1543825. Dividing ROIs by this value would
If you had 5.3 in addition to 4.5, you could run the following command
for each subject
cd $SUBJECTS_DIR/yoursubject
mri_segstats --seg mri/aseg.mgz --sum stats/aseg.53.stats --pv
mri/norm.mgz --empty --brainmask mri/brainmask.mgz --brain-vol-from-seg
--excludeid 0 --excl-ctxgmwm
The 5 condition analysis is not all that well conditioned either. I have
an (arbitrary) threshold of 10^6 max for the condition number. The
5-condition design is safely below the threshold, the 10-condition is
well above. Unfortunately, this threshold is hardwired. Do you really
need to do
Here is a paper discussing post-experiment power analyses and why they
are wrong:
http://amstat.tandfonline.com/doi/abs/10.1198/000313001300339897#.Um_aDtexGkc
On 10/29/2013 10:51 AM, MCLAREN, Donald wrote:
For anyone who wants G*Power, here is a link to the download page. It
is free for all
Hi suzanne, you'll need to use the 2mm version fsaverage/mri.2mm/aseg.mgz
doug
On 10/29/2013 11:46 AM, Suzanne Oosterwijk wrote:
Hello all,
I want to run a Monte Carlo simulation within a volume ROI and I am
running into a problem when I use the --mask flag while running
glmfit. My
Hi all,
I'm running analyses in Qdec and want to verify a few things:
1. When comparing two groups (HC vs patients) I am including mean cortical
thickness as a covariate, using the following equation:
bh.thickness = (lh.thickness*lh.surfarea + rh.thickness*rh.surfarea) /
(lh.surfarea +
On 10/29/2013 01:10 PM, Ashley Shurick wrote:
Hi all,
I'm running analyses in Qdec and want to verify a few things:
1. When comparing two groups (HC vs patients) I am including mean
cortical thickness as a covariate, using the following equation:
bh.thickness = (lh.thickness*lh.surfarea
19 matches
Mail list logo
