Hello, I want to run group statistics on the cortical thickness for a study with 3 groups, each with about 30 subjects. To do this, I need to create an average cort thickness map for each group. We are interested in basic statistics, so nothing too advanced. I am wondering what would be the most efficient way to go about this. From what I can tell, FsFast is the program used to do group analysis. Would it be fastest to learn FsFast to process all my group data, or I am thinking since we only need basic stats, it may be faster and easier to register each subject's ?h.thickness file to an average sphere and then load the registered thickness file into matlab to do the stats? I think I would need to modify the commands from auto-recon stage 3 to use the thickness file, for example:
mris_sphere ?h.thickness ?h.thickness.sphere mris_register -curv ?h.sphere $FREESURFER_HOME/average/?h.average.curvature.filled.buckner40.tif ?h.thickness.sphere.reg mris_jacobian ?h.white ?h.thickness.sphere.reg ?h.jacobian_white
mrisp_paint -a 5 $FREESURFER_HOME/average/?h.average.curvature.filled.buckner40.tif#6 ?h.thickness.sphere.reg ?h.thickness.avg_curvWe have a cluster, so the processing could be done fairly quickly. Thank you in advance for your advice!
Jeff Sadino
_________________________________________________________________ Stay up to date on your PC, the Web, and your mobile phone with Windows Live. http://clk.atdmt.com/MRT/go/msnnkwxp1020093185mrt/direct/01/
If this is just a thickness study, then you dont need fsfast. There are examples/tutorials of this type of analysis in the wiki. Look through that and see if you still have questions.
doug
Jeff Sadino wrote:
Hello, I want to run group statistics on the cortical thickness for a study with 3 groups, each with about 30 subjects. To do this, I need to create an average cort thickness map for each group. We are interested in basic statistics, so nothing too advanced. I am wondering what would be the most efficient way to go about this. From what I can tell, FsFast is the program used to do group analysis. Would it be fastest to learn FsFast to process all my group data, or I am thinking since we only need basic stats, it may be faster and easier to register each subject's ?h.thickness file to an average sphere and then load the registered thickness file into matlab to do the stats? I think I would need to modify the commands from auto-recon stage 3 to use the thickness file, for example:
mris_sphere ?h.thickness ?h.thickness.sphere mris_register http://surfer.nmr.mgh.harvard.edu/fswiki/mris_register -curv ?h.sphere $FREESURFER_HOME/average/?h.average.curvature.filled.buckner40.tif ?h.thickness.sphere.reg mris_jacobian http://surfer.nmr.mgh.harvard.edu/fswiki/mris_jacobian ?h.white ?h.thickness.sphere.reg ?h.jacobian_white
mrisp_paint http://surfer.nmr.mgh.harvard.edu/fswiki/mrisp_paint -a 5 $FREESURFER_HOME/average/?h.average.curvature.filled.buckner40.tif#6 ?h.thickness.sphere.reg ?h.thickness.avg_curv
We have a cluster, so the processing could be done fairly quickly. Thank you in advance for your advice!
Jeff Sadino
Stay up to date on your PC, the Web, and your mobile phone with Windows Live. See Now
http://clk.atdmt.com/MRT/go/msnnkwxp1020093185mrt/direct/01/
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
Thank you for your quick reply Doug. After looking, I think you are referring to the mri_glmfit tool? Is this the webpage to look at first to get started learning about how to do cortical thickness group analysis? http://surfer.nmr.mgh.harvard.edu/fswiki/FsTutorial/GroupAnalysis
Thank you, Jeff Sadino
Date: Wed, 18 Feb 2009 19:25:21 -0500 From: greve@nmr.mgh.harvard.edu To: jsadino@hotmail.com CC: freesurfer@nmr.mgh.harvard.edu Subject: Re: [Freesurfer] Group Cortical Thickness
If this is just a thickness study, then you dont need fsfast. There are examples/tutorials of this type of analysis in the wiki. Look through that and see if you still have questions.
doug
Jeff Sadino wrote:
Hello, I want to run group statistics on the cortical thickness for a study with 3 groups, each with about 30 subjects. To do this, I need to create an average cort thickness map for each group. We are interested in basic statistics, so nothing too advanced. I am wondering what would be the most efficient way to go about this. From what I can tell, FsFast is the program used to do group analysis. Would it be fastest to learn FsFast to process all my group data, or I am thinking since we only need basic stats, it may be faster and easier to register each subject's ?h.thickness file to an average sphere and then load the registered thickness file into matlab to do the stats? I think I would need to modify the commands from auto-recon stage 3 to use the thickness file, for example:
mris_sphere ?h.thickness ?h.thickness.sphere mris_register http://surfer.nmr.mgh.harvard.edu/fswiki/mris_register -curv ?h.sphere $FREESURFER_HOME/average/?h.average.curvature.filled.buckner40.tif ?h.thickness.sphere.reg mris_jacobian http://surfer.nmr.mgh.harvard.edu/fswiki/mris_jacobian ?h.white ?h.thickness.sphere.reg ?h.jacobian_white
mrisp_paint http://surfer.nmr.mgh.harvard.edu/fswiki/mrisp_paint -a 5 $FREESURFER_HOME/average/?h.average.curvature.filled.buckner40.tif#6 ?h.thickness.sphere.reg ?h.thickness.avg_curv
We have a cluster, so the processing could be done fairly quickly. Thank you in advance for your advice!
Jeff Sadino
Stay up to date on your PC, the Web, and your mobile phone with Windows Live. See Now
http://clk.atdmt.com/MRT/go/msnnkwxp1020093185mrt/direct/01/
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
_________________________________________________________________ Get more out of the Web. Learn 10 hidden secrets of Windows Live. http://windowslive.com/connect/post/jamiethomson.spaces.live.com-Blog-cns!55...
freesurfer@nmr.mgh.harvard.edu
-
Douglas N Greve -
Jeff Sadino