Hi, Doug Thank you for your answer. Well I guess that after I read your both tutorials for group analysis (http://surfer.nmr.mgh.harvard.edu/fswiki/FsTutorial/GroupAnalysis and http://surfer.nmr.mgh.harvard.edu/fswiki/FsTutorial/QdecGroupAnalysis ) I mistook the concepts of qdec.table.dat and FSGD. My fault.
Following the QDEC tutorial I saw we could import aseg data to qdec.table.dat after pressing "generate stats data tables" button. So I was wondering if I could do the same from hippo-subfields data. Importing is faster and more precise than make the table by hand.
Well the hippo-subfield pipeline creates several files such as posterior_left_subiculum.mgz on for each subfield and side. As far as I know you can: 1- run kvlQuantifyPosteriorProbabilityImages to output the volume in voxels 2- visualize the subfield segmentations with freeview loading nu.mgz and the posterior* files as p-labels.
Following QDEC tutorial we could generate a display showing the correlation between cortical thickness (or volume) and age for example. I would like to a similar analysis with hippo-subfields and get a display showing the correlations of the hippo-subfields and age. So, just for an example, perhaps I could notice that subiculum volume decreases with age but hippocampal fissure increases with age.
I guess I could analyze the data from hippo-subfields in a statistical package to get numbers and graphics but it would be great if I could see that as a surface like we do with cortical data. Cheers, Marcos Em Qui, 2013-05-16 às 11:56 -0400, Douglas N Greve escreveu:
Hi Marcos,
On 05/15/2013 08:32 PM, Marcos Martins da Silva wrote:
Hi, freesurfers I have two questions: 1- Is there a way to use QDEC to analyze data from hippocampal subfields created by recon-all -s subj -hippo-subfields? That is, I know I can manually put data from posterior* files when making the FSGD file but it would be great if I could import it like I do from aseg.stats.
when you say you want analyze the hippo subfields in QDEC, do you mean you want to include them as regressors?If so, you will need to add them to your QDEC table. BTW, QDEC does not needan FSGD file.
2- Assuming all data from my subjects are already are already in the FSGD, I understand I can select cortical thickness or volume as dependent variable and inspect the correlations between cortical thickness and age, for example. But I could not find how to define a hippocampal subfield volume as a dependent variable. Is that possible? This way I could display correlations between a subfield volume and age or gender.
I have not used the hippo subfields yet, and I'm embarrassed to say that I don't know what exactly it outputs. If there is a stats file there, you can run asegstats2table to create a table file, then use the table as input to mri_glmfit (use the --table option). doug
Thank you, Marcos
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer