External Email - Use Caution thank you for your detailed and helpful explanation.
question 3: I mean just left and right. But what should I do if I want to consider subcortical as well as left and right? Best Regards Hengameh On Thu, Mar 12, 2020 at 12:07 AM Douglas N. Greve <dgr...@mgh.harvard.edu> wrote: > > > On 3/11/2020 4:21 PM, Hengameh Marzbani wrote: > > External Email - Use Caution > > Thank you for the quick response. > > question 1: I need the results of mri-glmfit (before doing mri-glmfit-sim). I > can see the clusters by tksurfer. But, Can I have a text file included all > clusters? > > You can, the easiest way is to run mri_glmfit-sim and it will give you a > list of clusters in a text file. If you set the cwp to 1, then it will give > you all the clusters. See the help. > > question 3: I want to do the analyse by considering the whole brain. Is > correct to use --2spaces? and also I I want to visualize the results in both > hemisphere simultaneously. is it possible by freeview? > > Whole brain meaning left, right, and subcort or just left and right? If > just left and right, then use --2spaces > > question 5: I can't understand what is the reason of using demean and > normalize. Would you mind explaining to me? And when is it necessary to use > them? > > Demeaning makes the design matrix better conditioned. But also if you are > looking at the difference in average thickness between two groups > regressing out, eg, age, then simply using age as a regressor will reduce > the efficiency of the estimation because the age itself will have a > non-zero mean, so the regression is left trying to figure out what part of > the thickness average is due to the group and what part is due to age. > > Thank you so much. > > Hengameh > > > On Tue, Mar 10, 2020 at 2:03 AM Douglas N. Greve <dgr...@mgh.harvard.edu> > wrote: > >> >> >> On 3/9/20 5:53 PM, Hengameh Marzbani wrote: >> > >> > External Email - Use Caution >> > >> > Hello Freesurfer developers, >> > 1- I want to extract the uncorrected clusters from the uncorrected >> significance maps obtained from mri_glmfit for structural data (to >> illustrate cluster, size, location, and significance) like >> abs.sig.cluster.summary (summary of clusters (text))which is obtained from >> mri_glmfit-sim. Is there a way to do this? >> Set the clusterwise p-value thresh to 1 (--cwp 1) >> > 2- I don't see any clusters when I use cluster-based correction for >> multiple comparisons (for example mri_glmfit-sim \ --glmdir lh.Area.glmdir >> \ --sim mc-z 10000 2 mc-z.abs\ --sim-sign abs --cache 4 neg \ --cwp >> 0.05\ --2spaces). How much is it significant if I report the uncorrected >> clusters? >> Usually, reveiewers won't let you report exclusivly uncorrected results. >> > 3- Are there new ways of looking at the whole brain together (not left >> and right hemisphere separately)by Qdec and FSGD? >> You can load both hemispheres into freeview, if that is what you mean >> > 4- How can I do fdr correction in mri_glmfit-sim? what flag should I >> use? >> You can't, use mri_fdr. Run with --help to get docs >> > 5- How can I demean and normalize the covariates when I use FSGD? what >> flag should I use? >> Add these lines in your FSGD file >> DeMeanFlag 1 >> ReScaleFlag 1 >> > >> > Thanks you in advance! >> > Best regards, >> > Hengameh >> > >> > >> > >> > >> > _______________________________________________ >> > Freesurfer mailing list >> > Freesurfer@nmr.mgh.harvard.edu >> > https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer >> >> _______________________________________________ >> Freesurfer mailing list >> Freesurfer@nmr.mgh.harvard.edu >> https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer > > > _______________________________________________ > Freesurfer mailing > listfreesur...@nmr.mgh.harvard.eduhttps://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer > > > _______________________________________________ > Freesurfer mailing list > Freesurfer@nmr.mgh.harvard.edu > https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
_______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
