I am also very interested in these questions, but I do not find the response.
Could anyone provide the answers?
Thanks,
Feng
发件人："Meiling Li" <limeilingch...@163.com>
发送时间：2015-09-22 22:09
主题：[Freesurfer] GRF correction in mri_glmfit-sim
收件人："freesurfer-help"<freesurfer@nmr.mgh.harvard.edu>
抄送：
Dear FS experts
I do GRF correction using mri_glmfit-sim --grf in volume. I find --grf pos and
--grf neg, but not --grf abs.
My first question is, is the --grf just used for ONE-side correction? For
instance, we can test, are the FCs of patients larger than that of healthy
controls or smaller than the healthy controls? If I want to test the difference
of FCs between patients and healthy controls, how do i do that? Please help me
to check the following commands i used to do the TWO-sides correction (for
example. voxel-wise p: 0.05, cluster-wise p: 0.05):
mri_glmfit-sim --glmdir g2v0 --grf 1.3 pos --cwp .05
mri_glmfit-sim --glmdir g2v0 --grf 1.3 neg --cwp .05
Then i use the following commands to combine the 'pos' and 'neg' results
together
mri_convert grf.th1.3.neg.sig.cluster.mgh grf.th1.3.neg.sig.cluster.nii.gz
mri_convert grf.th1.3.pos.sig.cluster.mgh grf.th1.3.pos.sig.cluster.nii.gz
fslmaths grf.th1.3.pos.sig.cluster.nii.gz -add grf.th1.3.neg.sig.cluster.nii.gz
grf.th1.3.sig.cluster.nii.gz
Finally, i think the result grf.th1.3.sig.cluster.nii.gz is what i want, which
is the difference between patients and healthy controls with GRF corrected,
with vertex-wise p< 0.05 and cluster-wise p<0.05. Is it right?
2. The other question is, If i do the two-sides GRF correction(vertex-wise
p<0.05 and cluster-wise p<0.05) with 'pos' or 'neg', should the vertex-wise or
cluster-wise be 0.05/2 and 0.05/2? or 0.05 and 0.05?
for example, we should write like this
mri_glmfit-sim --glmdir g2v0 --grf -log10(0.05/2) pos --cwp 0.05/2
mri_glmfit-sim --glmdir g2v0 --grf -log10(0.05/2) neg --cwp 0.05/2
3. If i want to do cluster-based FDR correction, is there any FS command to do
it? is it right if i load the statistic p map into MATLAB and do the FDR
correction, then i use 'mri_volcluster --minsizevox' to threshold it to remove
voxels smaller than the cluster size i set.
Thank you very much for any help!
All the best,
Meiling
_______________________________________________
Freesurfer mailing list
Freesurfer@nmr.mgh.harvard.edu
https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
The information in this e-mail is intended only for the person to whom it is
addressed. If you believe this e-mail was sent to you in error and the e-mail
contains patient information, please contact the Partners Compliance HelpLine at
http://www.partners.org/complianceline . If the e-mail was sent to you in error
but does not contain patient information, please contact the sender and properly
dispose of the e-mail.
