Hi Suzanne, I'll cc the FS list so others can benefit ...
On 09/09/2013 06:02 AM, Suzanne Oosterwijk wrote:
Hi Doug,
I have two follow-up questions to enhance my own understanding of Freesurfer. I know that MC simulations take into account the smoothness of the data. I noticed that the smoothing factor of my surface maps is quite high (around 8/9 fwhm), even though I applied a smoothing factor of 5 during pre-processing. Can you explain to me why the final maps have such a high fwhm? Is this because of the transformations that are necessary to create the surfaces?
The final measured FWHM is a composite of the smoothing you applied plus any smoothness that may already be in the data. Eg, when the surfaces are created, there is a smoothness constraint that can cause smoothness in the surfaces. The gyrification index has very high smoothenss. For fMRI, there is already natural smoothness in the data plus smoothness added due to interpolation when sampling to the surface.
Another question concerns the relationship between voxels and vertices. Does one voxel corresponds to one vertex, or is the relationship different? This may be a very stupid question, but I thoroughly searched the wiki and the web without finding a straightforward answer.
Vertices and voxels are different entities. A voxel is a 3D box, a vertex is essentially a 2D triangle on the surface. The surface is made from a volume, of course. Initially, the square face of a voxel is divided into two triangles of equal size. The location of the vertices are then adjusted to get a better fit. Oftentimes, we use the term voxel and vertex interchangeably because each just represents a data point either on the surface or in the volume.
The next question is, if I want to apply a liberal threshold to my data that is similar to say p < .005; k > 20, then how do I produce this using the minarea flag in the mri_surfcluster command?
I'm not sure what you mean by p < .005. Is that a vertex-wise threshold (ie, the cluster-forming threshold) or cluster-wise? What is k? If you want to correct for multiple comparisons, then you can run mri_glmfit-sim. Once you select a cluster-wise and cluster-forming thresholds (and given the FWHM and search space), the critical clustersize can be determined.
doug
Thanks in advance for your answers! Suzanne
On Tue, Sep 3, 2013 at 3:57 PM, Douglas N Greve <greve@nmr.mgh.harvard.edu mailto:greve@nmr.mgh.harvard.edu> wrote:
Hi Suzanne, the pre-computed data is for a full hemisphere. if you have a smaller search space, then the whole hemi results will be too conservative. In that case, you should run mri_glmfit-sim without the --cache and with --sim as indicated in your command line (which looks correct to me). The simulations can take a while. If you send me your label, I can compute cached tables for you, so you don't have to keep running 10000 iterations of mri_glmfit-sim doug On 09/02/2013 12:04 PM, Suzanne Oosterwijk wrote: > [Attachment(s) <#TopText> from Suzanne Oosterwijk included below] > > Hi all, > > I have a question about the Monte Carlo simulation in Freesurfer. I > would like to run a simulation to define the minimum cluster size in a > specific collection of vertices (say within a ROI label). After > running a masked glmfit the pre-cached simulation doesn’t seem to > correct for the smaller number of vertices; the text files produce the > same results. I also ran a full simulation to see whether that made a > difference, and then I got different results. I just want to make sure > that this is because the smaller volume was applied and not simply > because the “real” simulation can differ from the pre-cached data. > Attached are the different output files, the cached MC with mask, the > cached MC without mask, the simulation with mask and a normal cluster > search within the mask (which demonstrates that there are > sub-threshold clusters). > > Below is my code, > > foreach s (f_fear_c_fear f_disg_c_disg f_mf_c_mf f_notc_c_notc) > > pushd ${s} > > mri_glmfit --surf fsaverage lh --y ces.nii.gz --osgm --glmdir > glm.masked --label > /home/sooster1/Desktop/FALSE_FEEDBACK_imaging/DATA/feedback_structural/network_labels/FP_DN_network.label > > mri_glmfit-sim --cache 2.3 pos --sim-sign pos --cwpvalthresh .05 > --glmdir glm.masked > > mri_surfcluster --sign pos --hemi lh --in glm.masked/osgm/sig.mgh > --thmin 2.3 --thmax infinity --minarea 50 --subject fsaverage --sum > glm.masked/osgm/cluster_pos_23.sum --o > ./glm.masked/osgm/sig.cluster_pos_23.mgh --annot aparc > > cd .. > > end > > > I used this code to run the simulation: > > mri_glmfit-sim --glmdir glm.masked --sim mc-z 10000 2.3 mc-z.pos.23 > --sim-sign pos > > Thanks in advance for responding! > > Suzanne > -- Douglas N. Greve, Ph.D. MGH-NMR Center greve@nmr.mgh.harvard.edu <mailto:greve@nmr.mgh.harvard.edu> Phone Number: 617-724-2358 <tel:617-724-2358> Fax: 617-726-7422 <tel:617-726-7422> Bugs: surfer.nmr.mgh.harvard.edu/fswiki/BugReporting <http://surfer.nmr.mgh.harvard.edu/fswiki/BugReporting> FileDrop: https://gate.nmr.mgh.harvard.edu/filedrop2 www.nmr.mgh.harvard.edu/facility/filedrop/index.html <http://www.nmr.mgh.harvard.edu/facility/filedrop/index.html> Outgoing: ftp://surfer.nmr.mgh.harvard.edu/transfer/outgoing/flat/greve/ 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. ------------------------------------ Yahoo! Groups Links <*> To visit your group on the web, go to: http://groups.yahoo.com/group/martinos-tech/ <*> Your email settings: Individual Email | Traditional <*> To change settings online go to: http://groups.yahoo.com/group/martinos-tech/join (Yahoo! ID required) <*> To change settings via email: martinos-tech-digest@yahoogroups.com <mailto:martinos-tech-digest@yahoogroups.com> martinos-tech-fullfeatured@yahoogroups.com <mailto:martinos-tech-fullfeatured@yahoogroups.com> <*> To unsubscribe from this group, send an email to: martinos-tech-unsubscribe@yahoogroups.com <mailto:martinos-tech-unsubscribe@yahoogroups.com> <*> Your use of Yahoo! Groups is subject to: http://docs.yahoo.com/info/terms/
Hi Doug,
Thanks for your reply. I know how to run mri_glmfit-sim. What I'd like to do is to apply a more liberal threshold to my data to examine the clusters that fall outside of the MC-corrected threshold. To do this I applied the mri_surfcluster command with a vertex-wise threshold of p < .005, and a minarea of 50 mm2. My question is how I should interpret the minarea flag. I want to have some insight in whether this threshold is more liberal than a threshold of 20 continuous voxels with a p-value of .005 (p < .005; k > 20) or whether this threshold is more conservative.
Thanks, Suzanne
On Wed, Sep 11, 2013 at 12:08 AM, Douglas N Greve <greve@nmr.mgh.harvard.edu
wrote:
Hi Suzanne, I'll cc the FS list so others can benefit ...
On 09/09/2013 06:02 AM, Suzanne Oosterwijk wrote:
Hi Doug,
I have two follow-up questions to enhance my own understanding of Freesurfer. I know that MC simulations take into account the smoothness of the data. I noticed that the smoothing factor of my surface maps is quite high (around 8/9 fwhm), even though I applied a smoothing factor of 5 during pre-processing. Can you explain to me why the final maps have such a high fwhm? Is this because of the transformations that are necessary to create the surfaces?
The final measured FWHM is a composite of the smoothing you applied plus any smoothness that may already be in the data. Eg, when the surfaces are created, there is a smoothness constraint that can cause smoothness in the surfaces. The gyrification index has very high smoothenss. For fMRI, there is already natural smoothness in the data plus smoothness added due to interpolation when sampling to the surface.
Another question concerns the relationship between voxels and vertices. Does one voxel corresponds to one vertex, or is the relationship different? This may be a very stupid question, but I thoroughly searched the wiki and the web without finding a straightforward answer.
Vertices and voxels are different entities. A voxel is a 3D box, a vertex is essentially a 2D triangle on the surface. The surface is made from a volume, of course. Initially, the square face of a voxel is divided into two triangles of equal size. The location of the vertices are then adjusted to get a better fit. Oftentimes, we use the term voxel and vertex interchangeably because each just represents a data point either on the surface or in the volume.
The next question is, if I want to apply a liberal threshold to my data
that is similar to say p < .005; k > 20, then how do I produce this using the minarea flag in the mri_surfcluster command?
I'm not sure what you mean by p < .005. Is that a vertex-wise threshold (ie, the cluster-forming threshold) or cluster-wise? What is k? If you want to correct for multiple comparisons, then you can run mri_glmfit-sim. Once you select a cluster-wise and cluster-forming thresholds (and given the FWHM and search space), the critical clustersize can be determined.
doug
Thanks in advance for your answers! Suzanne
On Tue, Sep 3, 2013 at 3:57 PM, Douglas N Greve < greve@nmr.mgh.harvard.edu <mailto:greve@nmr.mgh.harvard.**edugreve@nmr.mgh.harvard.edu>> wrote:
Hi Suzanne, the pre-computed data is for a full hemisphere. if you have a smaller search space, then the whole hemi results will be too conservative. In that case, you should run mri_glmfit-sim without the --cache and with --sim as indicated in your command line (which looks correct to me). The simulations can take a while. If you send me your label, I can compute cached tables for you, so you don't have to keep running 10000 iterations of mri_glmfit-sim doug On 09/02/2013 12:04 PM, Suzanne Oosterwijk wrote: > [Attachment(s) <#TopText> from Suzanne Oosterwijk included below] > > Hi all, > > I have a question about the Monte Carlo simulation in Freesurfer. I > would like to run a simulation to define the minimum cluster size in a > specific collection of vertices (say within a ROI label). After > running a masked glmfit the pre-cached simulation doesn’t seem to > correct for the smaller number of vertices; the text files produce the > same results. I also ran a full simulation to see whether that made a > difference, and then I got different results. I just want to make sure > that this is because the smaller volume was applied and not simply > because the “real” simulation can differ from the pre-cached data. > Attached are the different output files, the cached MC with mask, the > cached MC without mask, the simulation with mask and a normal cluster > search within the mask (which demonstrates that there are > sub-threshold clusters). > > Below is my code, > > foreach s (f_fear_c_fear f_disg_c_disg f_mf_c_mf f_notc_c_notc) > > pushd ${s} > > mri_glmfit --surf fsaverage lh --y ces.nii.gz --osgm --glmdir > glm.masked --label > /home/sooster1/Desktop/FALSE_**FEEDBACK_imaging/DATA/**feedback_structural/network_**labels/FP_DN_network.label > > mri_glmfit-sim --cache 2.3 pos --sim-sign pos --cwpvalthresh .05 > --glmdir glm.masked > > mri_surfcluster --sign pos --hemi lh --in glm.masked/osgm/sig.mgh > --thmin 2.3 --thmax infinity --minarea 50 --subject fsaverage --sum > glm.masked/osgm/cluster_pos_**23.sum --o > ./glm.masked/osgm/sig.cluster_**pos_23.mgh --annot aparc > > cd .. > > end > > > I used this code to run the simulation: > > mri_glmfit-sim --glmdir glm.masked --sim mc-z 10000 2.3 mc-z.pos.23 > --sim-sign pos > > Thanks in advance for responding! > > Suzanne >
-- Douglas N. Greve, Ph.D. MGH-NMR Center greve@nmr.mgh.harvard.edu <mailto:greve@nmr.mgh.harvard.**edu<greve@nmr.mgh.harvard.edu>Phone Number: 617-724-2358 <tel:617-724-2358> Fax: 617-726-7422 <tel:617-726-7422> Bugs: surfer.nmr.mgh.harvard.edu/**fswiki/BugReporting<http://surfer.nmr.mgh.harvard.edu/fswiki/BugReporting> <http://surfer.nmr.mgh.**harvard.edu/fswiki/**BugReporting<http://surfer.nmr.mgh.harvard.edu/fswiki/BugReporting>FileDrop: https://gate.nmr.mgh.harvard.**edu/filedrop2<https://gate.nmr.mgh.harvard.edu/filedrop2> www.nmr.mgh.harvard.edu/**facility/filedrop/index.html<http://www.nmr.mgh.harvard.edu/facility/filedrop/index.html> <http://www.nmr.mgh.harvard.**edu/facility/filedrop/index.**html<http://www.nmr.mgh.harvard.edu/facility/filedrop/index.html>Outgoing: ftp://surfer.nmr.mgh.harvard.**edu/transfer/outgoing/flat/**greve/<ftp://surfer.nmr.mgh.harvard.edu/transfer/outgoing/flat/greve/> 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<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. ------------------------------**------ Yahoo! Groups Links <*> To visit your group on the web, go to: http://groups.yahoo.com/group/**martinos-tech/<http://groups.yahoo.com/group/martinos-tech/> <*> Your email settings: Individual Email | Traditional <*> To change settings online go to: http://groups.yahoo.com/group/**martinos-tech/join<http://groups.yahoo.com/group/martinos-tech/join> (Yahoo! ID required) <*> To change settings via email: martinos-tech-digest@**yahoogroups.com<martinos-tech-digest@yahoogroups.com> <mailto:martinos-tech-digest@**yahoogroups.com<martinos-tech-digest@yahoogroups.com>martinos-tech-fullfeatured@**yahoogroups.com<martinos-tech-fullfeatured@yahoogroups.com> <mailto:martinos-tech-**fullfeatured@yahoogroups.com<martinos-tech-fullfeatured@yahoogroups.com><*> To unsubscribe from this group, send an email to: martinos-tech-unsubscribe@**yahoogroups.com<martinos-tech-unsubscribe@yahoogroups.com> <mailto:martinos-tech-**unsubscribe@yahoogroups.com<martinos-tech-unsubscribe@yahoogroups.com><*> Your use of Yahoo! Groups is subject to: http://docs.yahoo.com/info/**terms/<http://docs.yahoo.com/info/terms/>-- Douglas N. Greve, Ph.D. MGH-NMR Center greve@nmr.mgh.harvard.edu Phone Number: 617-724-2358 Fax: 617-726-7422
Bugs: surfer.nmr.mgh.harvard.edu/**fswiki/BugReportinghttp://surfer.nmr.mgh.harvard.edu/fswiki/BugReporting FileDrop: https://gate.nmr.mgh.harvard.**edu/filedrop2https://gate.nmr.mgh.harvard.edu/filedrop2 www.nmr.mgh.harvard.edu/**facility/filedrop/index.htmlhttp://www.nmr.mgh.harvard.edu/facility/filedrop/index.html Outgoing: ftp://surfer.nmr.mgh.harvard.**edu/transfer/outgoing/flat/** greve/ ftp://surfer.nmr.mgh.harvard.edu/transfer/outgoing/flat/greve/
Suzanne,
p<0.005 is the same whether its voxels or vertices. The liberal nature of 20 voxels is vague to begin with as there is two things contributing to how liberal 20 voxels are: (1) voxel size - 20 1mm isotropic voxels is quite different from 20 3mm isotropic voxels; (2) The number of voxels you are testing -- testing 100 voxels and finding 20 that are significant is quite different than testing 100000 voxels and finding 20 that are significant.
Thus, trying to compare voxels to surface area is quite difficult. I am not sure if there is a way to equate the surface area to voxels given points #1 and #2. It's hard to equate the number of voxels to the surface area because voxels occur outside of the surface as well. You'd have to constrain your analysis to only grey matter and p <0.005 in 20 voxels to be testing the same measurements. Even then, the tests wouldn't quite be the same number or areas.
Best Regards, Donald McLaren ================= D.G. McLaren, Ph.D. Research Fellow, Department of Neurology, Massachusetts General Hospital and Harvard Medical School Postdoctoral Research Fellow, GRECC, Bedford VA Website: http://www.martinos.org/~mclaren Office: (773) 406-2464 ===================== This e-mail contains CONFIDENTIAL INFORMATION which may contain PROTECTED HEALTHCARE INFORMATION and may also be LEGALLY PRIVILEGED and which is intended only for the use of the individual or entity named above. If the reader of the e-mail is not the intended recipient or the employee or agent responsible for delivering it to the intended recipient, you are hereby notified that you are in possession of confidential and privileged information. Any unauthorized use, disclosure, copying or the taking of any action in reliance on the contents of this information is strictly prohibited and may be unlawful. If you have received this e-mail unintentionally, please immediately notify the sender via telephone at (773) 406-2464 or email.
On Wed, Sep 11, 2013 at 4:20 AM, Suzanne Oosterwijk < suzanne.oosterwijk@gmail.com> wrote:
Hi Doug,
Thanks for your reply. I know how to run mri_glmfit-sim. What I'd like to do is to apply a more liberal threshold to my data to examine the clusters that fall outside of the MC-corrected threshold. To do this I applied the mri_surfcluster command with a vertex-wise threshold of p < .005, and a minarea of 50 mm2. My question is how I should interpret the minarea flag. I want to have some insight in whether this threshold is more liberal than a threshold of 20 continuous voxels with a p-value of .005 (p < .005; k > 20) or whether this threshold is more conservative.
Thanks, Suzanne
On Wed, Sep 11, 2013 at 12:08 AM, Douglas N Greve < greve@nmr.mgh.harvard.edu> wrote:
Hi Suzanne, I'll cc the FS list so others can benefit ...
On 09/09/2013 06:02 AM, Suzanne Oosterwijk wrote:
Hi Doug,
I have two follow-up questions to enhance my own understanding of Freesurfer. I know that MC simulations take into account the smoothness of the data. I noticed that the smoothing factor of my surface maps is quite high (around 8/9 fwhm), even though I applied a smoothing factor of 5 during pre-processing. Can you explain to me why the final maps have such a high fwhm? Is this because of the transformations that are necessary to create the surfaces?
The final measured FWHM is a composite of the smoothing you applied plus any smoothness that may already be in the data. Eg, when the surfaces are created, there is a smoothness constraint that can cause smoothness in the surfaces. The gyrification index has very high smoothenss. For fMRI, there is already natural smoothness in the data plus smoothness added due to interpolation when sampling to the surface.
Another question concerns the relationship between voxels and vertices. Does one voxel corresponds to one vertex, or is the relationship different? This may be a very stupid question, but I thoroughly searched the wiki and the web without finding a straightforward answer.
Vertices and voxels are different entities. A voxel is a 3D box, a vertex is essentially a 2D triangle on the surface. The surface is made from a volume, of course. Initially, the square face of a voxel is divided into two triangles of equal size. The location of the vertices are then adjusted to get a better fit. Oftentimes, we use the term voxel and vertex interchangeably because each just represents a data point either on the surface or in the volume.
The next question is, if I want to apply a liberal threshold to my data
that is similar to say p < .005; k > 20, then how do I produce this using the minarea flag in the mri_surfcluster command?
I'm not sure what you mean by p < .005. Is that a vertex-wise threshold (ie, the cluster-forming threshold) or cluster-wise? What is k? If you want to correct for multiple comparisons, then you can run mri_glmfit-sim. Once you select a cluster-wise and cluster-forming thresholds (and given the FWHM and search space), the critical clustersize can be determined.
doug
Thanks in advance for your answers! Suzanne
On Tue, Sep 3, 2013 at 3:57 PM, Douglas N Greve < greve@nmr.mgh.harvard.edu <mailto:greve@nmr.mgh.harvard.**edugreve@nmr.mgh.harvard.edu>> wrote:
Hi Suzanne, the pre-computed data is for a full hemisphere. if you have a smaller search space, then the whole hemi results will be too conservative. In that case, you should run mri_glmfit-sim without the --cache and with --sim as indicated in your command line (which looks correct to me). The simulations can take a while. If you send me your label, I can compute cached tables for you, so you don't have to keep running 10000 iterations of mri_glmfit-sim doug On 09/02/2013 12:04 PM, Suzanne Oosterwijk wrote: > [Attachment(s) <#TopText> from Suzanne Oosterwijk included below] > > Hi all, > > I have a question about the Monte Carlo simulation in Freesurfer. I > would like to run a simulation to define the minimum cluster size in a > specific collection of vertices (say within a ROI label). After > running a masked glmfit the pre-cached simulation doesn’t seem to > correct for the smaller number of vertices; the text files produce the > same results. I also ran a full simulation to see whether that made a > difference, and then I got different results. I just want to make sure > that this is because the smaller volume was applied and not simply > because the “real” simulation can differ from the pre-cached data. > Attached are the different output files, the cached MC with mask, the > cached MC without mask, the simulation with mask and a normal cluster > search within the mask (which demonstrates that there are > sub-threshold clusters). > > Below is my code, > > foreach s (f_fear_c_fear f_disg_c_disg f_mf_c_mf f_notc_c_notc) > > pushd ${s} > > mri_glmfit --surf fsaverage lh --y ces.nii.gz --osgm --glmdir > glm.masked --label > /home/sooster1/Desktop/FALSE_**FEEDBACK_imaging/DATA/**feedback_structural/network_**labels/FP_DN_network.label > > mri_glmfit-sim --cache 2.3 pos --sim-sign pos --cwpvalthresh .05 > --glmdir glm.masked > > mri_surfcluster --sign pos --hemi lh --in glm.masked/osgm/sig.mgh > --thmin 2.3 --thmax infinity --minarea 50 --subject fsaverage --sum > glm.masked/osgm/cluster_pos_**23.sum --o > ./glm.masked/osgm/sig.cluster_**pos_23.mgh --annot aparc > > cd .. > > end > > > I used this code to run the simulation: > > mri_glmfit-sim --glmdir glm.masked --sim mc-z 10000 2.3 mc-z.pos.23 > --sim-sign pos > > Thanks in advance for responding! > > Suzanne >
-- Douglas N. Greve, Ph.D. MGH-NMR Center greve@nmr.mgh.harvard.edu <mailto:greve@nmr.mgh.harvard.**edu<greve@nmr.mgh.harvard.edu>Phone Number: 617-724-2358 <tel:617-724-2358> Fax: 617-726-7422 <tel:617-726-7422> Bugs: surfer.nmr.mgh.harvard.edu/**fswiki/BugReporting<http://surfer.nmr.mgh.harvard.edu/fswiki/BugReporting> <http://surfer.nmr.mgh.**harvard.edu/fswiki/**BugReporting<http://surfer.nmr.mgh.harvard.edu/fswiki/BugReporting>FileDrop: https://gate.nmr.mgh.harvard.**edu/filedrop2<https://gate.nmr.mgh.harvard.edu/filedrop2> www.nmr.mgh.harvard.edu/**facility/filedrop/index.html<http://www.nmr.mgh.harvard.edu/facility/filedrop/index.html> <http://www.nmr.mgh.harvard.**edu/facility/filedrop/index.**html<http://www.nmr.mgh.harvard.edu/facility/filedrop/index.html>Outgoing: ftp://surfer.nmr.mgh.harvard.**edu/transfer/outgoing/flat/**greve/<ftp://surfer.nmr.mgh.harvard.edu/transfer/outgoing/flat/greve/> 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<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. ------------------------------**------ Yahoo! Groups Links <*> To visit your group on the web, go to: http://groups.yahoo.com/group/**martinos-tech/<http://groups.yahoo.com/group/martinos-tech/> <*> Your email settings: Individual Email | Traditional <*> To change settings online go to: http://groups.yahoo.com/group/**martinos-tech/join<http://groups.yahoo.com/group/martinos-tech/join> (Yahoo! ID required) <*> To change settings via email: martinos-tech-digest@**yahoogroups.com<martinos-tech-digest@yahoogroups.com> <mailto:martinos-tech-digest@**yahoogroups.com<martinos-tech-digest@yahoogroups.com>martinos-tech-fullfeatured@**yahoogroups.com<martinos-tech-fullfeatured@yahoogroups.com> <mailto:martinos-tech-**fullfeatured@yahoogroups.com<martinos-tech-fullfeatured@yahoogroups.com><*> To unsubscribe from this group, send an email to: martinos-tech-unsubscribe@**yahoogroups.com<martinos-tech-unsubscribe@yahoogroups.com> <mailto:martinos-tech-**unsubscribe@yahoogroups.com<martinos-tech-unsubscribe@yahoogroups.com><*> Your use of Yahoo! Groups is subject to: http://docs.yahoo.com/info/**terms/<http://docs.yahoo.com/info/terms/>-- Douglas N. Greve, Ph.D. MGH-NMR Center greve@nmr.mgh.harvard.edu Phone Number: 617-724-2358 Fax: 617-726-7422
Bugs: surfer.nmr.mgh.harvard.edu/**fswiki/BugReportinghttp://surfer.nmr.mgh.harvard.edu/fswiki/BugReporting FileDrop: https://gate.nmr.mgh.harvard.**edu/filedrop2https://gate.nmr.mgh.harvard.edu/filedrop2 www.nmr.mgh.harvard.edu/**facility/filedrop/index.htmlhttp://www.nmr.mgh.harvard.edu/facility/filedrop/index.html Outgoing: ftp://surfer.nmr.mgh.harvard.**edu/transfer/outgoing/flat/** greve/ ftp://surfer.nmr.mgh.harvard.edu/transfer/outgoing/flat/greve/
--
Suzanne Oosterwijk, Ph.D.
Postdoctoral Researcher
Department of Social Psychology
University of Amsterdam
s.oosterwijk@u s.oosterwijk@neu.eduva.nl
https://sites.google.com/site/suzanneoosterwijk/
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.
Hi Suzanne, why would you not just re-run mri_glmfit-sim with the different vertex wise threshold? If you want to use 20 voxels, then you'd have to look a the table and see what cluster p-value that corresponds to with vertex threshold = .005 doug
On 9/11/13 10:20 AM, Suzanne Oosterwijk wrote:
Hi Doug,
Thanks for your reply. I know how to run mri_glmfit-sim. What I'd like to do is to apply a more liberal threshold to my data to examine the clusters that fall outside of the MC-corrected threshold. To do this I applied the mri_surfcluster command with a vertex-wise threshold of p < .005, and a minarea of 50 mm2. My question is how I should interpret the minarea flag. I want to have some insight in whether this threshold is more liberal than a threshold of 20 continuous voxels with a p-value of .005 (p < .005; k > 20) or whether this threshold is more conservative.
Thanks, Suzanne
On Wed, Sep 11, 2013 at 12:08 AM, Douglas N Greve <greve@nmr.mgh.harvard.edu mailto:greve@nmr.mgh.harvard.edu> wrote:
Hi Suzanne, I'll cc the FS list so others can benefit ... On 09/09/2013 06:02 AM, Suzanne Oosterwijk wrote: Hi Doug, I have two follow-up questions to enhance my own understanding of Freesurfer. I know that MC simulations take into account the smoothness of the data. I noticed that the smoothing factor of my surface maps is quite high (around 8/9 fwhm), even though I applied a smoothing factor of 5 during pre-processing. Can you explain to me why the final maps have such a high fwhm? Is this because of the transformations that are necessary to create the surfaces? The final measured FWHM is a composite of the smoothing you applied plus any smoothness that may already be in the data. Eg, when the surfaces are created, there is a smoothness constraint that can cause smoothness in the surfaces. The gyrification index has very high smoothenss. For fMRI, there is already natural smoothness in the data plus smoothness added due to interpolation when sampling to the surface. Another question concerns the relationship between voxels and vertices. Does one voxel corresponds to one vertex, or is the relationship different? This may be a very stupid question, but I thoroughly searched the wiki and the web without finding a straightforward answer. Vertices and voxels are different entities. A voxel is a 3D box, a vertex is essentially a 2D triangle on the surface. The surface is made from a volume, of course. Initially, the square face of a voxel is divided into two triangles of equal size. The location of the vertices are then adjusted to get a better fit. Oftentimes, we use the term voxel and vertex interchangeably because each just represents a data point either on the surface or in the volume. The next question is, if I want to apply a liberal threshold to my data that is similar to say p < .005; k > 20, then how do I produce this using the minarea flag in the mri_surfcluster command? I'm not sure what you mean by p < .005. Is that a vertex-wise threshold (ie, the cluster-forming threshold) or cluster-wise? What is k? If you want to correct for multiple comparisons, then you can run mri_glmfit-sim. Once you select a cluster-wise and cluster-forming thresholds (and given the FWHM and search space), the critical clustersize can be determined. doug Thanks in advance for your answers! Suzanne On Tue, Sep 3, 2013 at 3:57 PM, Douglas N Greve <greve@nmr.mgh.harvard.edu <mailto:greve@nmr.mgh.harvard.edu> <mailto:greve@nmr.mgh.harvard.edu <mailto:greve@nmr.mgh.harvard.edu>>> wrote: Hi Suzanne, the pre-computed data is for a full hemisphere. if you have a smaller search space, then the whole hemi results will be too conservative. In that case, you should run mri_glmfit-sim without the --cache and with --sim as indicated in your command line (which looks correct to me). The simulations can take a while. If you send me your label, I can compute cached tables for you, so you don't have to keep running 10000 iterations of mri_glmfit-sim doug On 09/02/2013 12:04 PM, Suzanne Oosterwijk wrote: > [Attachment(s) <#TopText> from Suzanne Oosterwijk included below] > > Hi all, > > I have a question about the Monte Carlo simulation in Freesurfer. I > would like to run a simulation to define the minimum cluster size in a > specific collection of vertices (say within a ROI label). After > running a masked glmfit the pre-cached simulation doesn’t seem to > correct for the smaller number of vertices; the text files produce the > same results. I also ran a full simulation to see whether that made a > difference, and then I got different results. I just want to make sure > that this is because the smaller volume was applied and not simply > because the “real” simulation can differ from the pre-cached data. > Attached are the different output files, the cached MC with mask, the > cached MC without mask, the simulation with mask and a normal cluster > search within the mask (which demonstrates that there are > sub-threshold clusters). > > Below is my code, > > foreach s (f_fear_c_fear f_disg_c_disg f_mf_c_mf f_notc_c_notc) > > pushd ${s} > > mri_glmfit --surf fsaverage lh --y ces.nii.gz --osgm --glmdir > glm.masked --label > /home/sooster1/Desktop/FALSE_FEEDBACK_imaging/DATA/feedback_structural/network_labels/FP_DN_network.label > > mri_glmfit-sim --cache 2.3 pos --sim-sign pos --cwpvalthresh .05 > --glmdir glm.masked > > mri_surfcluster --sign pos --hemi lh --in glm.masked/osgm/sig.mgh > --thmin 2.3 --thmax infinity --minarea 50 --subject fsaverage --sum > glm.masked/osgm/cluster_pos_23.sum --o > ./glm.masked/osgm/sig.cluster_pos_23.mgh --annot aparc > > cd .. > > end > > > I used this code to run the simulation: > > mri_glmfit-sim --glmdir glm.masked --sim mc-z 10000 2.3 mc-z.pos.23 > --sim-sign pos > > Thanks in advance for responding! > > Suzanne > -- Douglas N. Greve, Ph.D. MGH-NMR Center greve@nmr.mgh.harvard.edu <mailto:greve@nmr.mgh.harvard.edu> <mailto:greve@nmr.mgh.harvard.edu <mailto:greve@nmr.mgh.harvard.edu>> Phone Number: 617-724-2358 <tel:617-724-2358> <tel:617-724-2358 <tel:617-724-2358>> Fax: 617-726-7422 <tel:617-726-7422> <tel:617-726-7422 <tel:617-726-7422>> Bugs: surfer.nmr.mgh.harvard.edu/fswiki/BugReporting <http://surfer.nmr.mgh.harvard.edu/fswiki/BugReporting> <http://surfer.nmr.mgh.harvard.edu/fswiki/BugReporting> FileDrop: https://gate.nmr.mgh.harvard.edu/filedrop2 www.nmr.mgh.harvard.edu/facility/filedrop/index.html <http://www.nmr.mgh.harvard.edu/facility/filedrop/index.html> <http://www.nmr.mgh.harvard.edu/facility/filedrop/index.html> Outgoing: ftp://surfer.nmr.mgh.harvard.edu/transfer/outgoing/flat/greve/ 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. ------------------------------------ Yahoo! Groups Links <*> To visit your group on the web, go to: http://groups.yahoo.com/group/martinos-tech/ <*> Your email settings: Individual Email | Traditional <*> To change settings online go to: http://groups.yahoo.com/group/martinos-tech/join (Yahoo! ID required) <*> To change settings via email: martinos-tech-digest@yahoogroups.com <mailto:martinos-tech-digest@yahoogroups.com> <mailto:martinos-tech-digest@yahoogroups.com <mailto:martinos-tech-digest@yahoogroups.com>> martinos-tech-fullfeatured@yahoogroups.com <mailto:martinos-tech-fullfeatured@yahoogroups.com> <mailto:martinos-tech-fullfeatured@yahoogroups.com <mailto:martinos-tech-fullfeatured@yahoogroups.com>> <*> To unsubscribe from this group, send an email to: martinos-tech-unsubscribe@yahoogroups.com <mailto:martinos-tech-unsubscribe@yahoogroups.com> <mailto:martinos-tech-unsubscribe@yahoogroups.com <mailto:martinos-tech-unsubscribe@yahoogroups.com>> <*> Your use of Yahoo! Groups is subject to: http://docs.yahoo.com/info/terms/ -- Douglas N. Greve, Ph.D. MGH-NMR Center greve@nmr.mgh.harvard.edu <mailto:greve@nmr.mgh.harvard.edu> Phone Number: 617-724-2358 <tel:617-724-2358> Fax: 617-726-7422 <tel:617-726-7422> Bugs: surfer.nmr.mgh.harvard.edu/fswiki/BugReporting <http://surfer.nmr.mgh.harvard.edu/fswiki/BugReporting> FileDrop: https://gate.nmr.mgh.harvard.edu/filedrop2 www.nmr.mgh.harvard.edu/facility/filedrop/index.html <http://www.nmr.mgh.harvard.edu/facility/filedrop/index.html> Outgoing: ftp://surfer.nmr.mgh.harvard.edu/transfer/outgoing/flat/greve/--
Suzanne Oosterwijk, Ph.D.
Postdoctoral Researcher
Department of Social Psychology
University of Amsterdam
s.oosterwijk@u mailto:s.oosterwijk@neu.eduva.nl http://va.nl
freesurfer@nmr.mgh.harvard.edu
-
Douglas Greve -
Douglas N Greve
-
MCLAREN, Donald -
Suzanne Oosterwijk