Dear Freesurfer experts,
I have a quick question regarding the appropriate design (DOSS or DODS) to be used with QDEC when comparing 2 groups for cortical thickness and LGI values.
So far I have used DODS first to look for an age-cortical thickness interaction in both groups. I used the real ages and also demeaned the age variable in my analysis. I entered age as a covariate and as a nuisance factor also. I have gender and diagnosis as discrete and eTIV as nuisance in both analysis. Neither was significant.
After this I used DOSS in QDEC to compare groups groups for both cortical thickness and LGI.
My questions relates to the use of age as a covariate in DOSS. Is it appropriate to enter age as covariate in DOSS given that DOSS assumes that both groups have the same slope? I have looked through the archives and online for an answer to this but I am still a bit confused.
Any help or advice on this would be greatly appreciated.
Thank you so much in advance.
Best wishes
Marco
DOSS should not be used with QDEC as it may or may not do the right thing. Is it not disabled? What version are you using?
To answer your question, if you want to use DOSS and you're interested in the diff bet groups, then first test for an interaction between group and covariate. If that is not significant, then use DOSS
On 6/20/17 5:05 AM, marco mcsweeney wrote:
Dear Freesurfer experts,
I have a quick question regarding the appropriate design (DOSS or DODS) to be used with QDEC when comparing 2 groups for cortical thickness and LGI values.
So far I have used DODS first to look for an age-cortical thickness interaction in both groups. I used the real ages and also demeaned the age variable in my analysis. I entered age as a covariate and as a nuisance factor also. I have gender and diagnosis as discrete and eTIV as nuisance in both analysis. Neither was significant.
After this I used DOSS in QDEC to compare groups groups for both cortical thickness and LGI.
My questions relates to the use of age as a covariate in DOSS. Is it appropriate to enter age as covariate in DOSS given that DOSS assumes that both groups have the same slope? I have looked through the archives and online for an answer to this but I am still a bit confused.
Any help or advice on this would be greatly appreciated.
Thank you so much in advance.
Best wishes
Marco
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
Dear Doug,
thank you so much for getting back to me on this.
I am using Freesurfer version 5.3 with QDEV version 1.4 which gives me both DODS and DOSS design matrix options. I am a little confused about DOSS in QDEC as I have read quite a few publications which have used DOSS in QDEC before. Can you please let me know what the issue with DOSS in QDEC is?
Also, I have looked at age-cortical thickness interaction using DODS in Qdec with gender and diagnosis as fixed factors, age as covariate (demean age as well) and eTIV as nuisance. I did not find any difference in age-cortical thickness interaction between the 2 groups.
Also, I am still a bit unsure about using age as covariate in DOSS through the command line. Is this appropriate given that DOSS assumes the same slope in both groups? Or should age be used as a nuisance factor instead (0 in the design matrix)?
Thank you so much again for your help with this.
Best wishes
Marco
On Tue, Jun 20, 2017 at 2:02 PM, Douglas Greve greve@nmr.mgh.harvard.edu wrote:
DOSS should not be used with QDEC as it may or may not do the right thing. Is it not disabled? What version are you using?
To answer your question, if you want to use DOSS and you're interested in the diff bet groups, then first test for an interaction between group and covariate. If that is not significant, then use DOSS
On 6/20/17 5:05 AM, marco mcsweeney wrote:
Dear Freesurfer experts,
I have a quick question regarding the appropriate design (DOSS or DODS) to be used with QDEC when comparing 2 groups for cortical thickness and LGI values.
So far I have used DODS first to look for an age-cortical thickness interaction in both groups. I used the real ages and also demeaned the age variable in my analysis. I entered age as a covariate and as a nuisance factor also. I have gender and diagnosis as discrete and eTIV as nuisance in both analysis. Neither was significant.
After this I used DOSS in QDEC to compare groups groups for both cortical thickness and LGI.
My questions relates to the use of age as a covariate in DOSS. Is it appropriate to enter age as covariate in DOSS given that DOSS assumes that both groups have the same slope? I have looked through the archives and online for an answer to this but I am still a bit confused.
Any help or advice on this would be greatly appreciated.
Thank you so much in advance.
Best wishes
Marco
Freesurfer mailing listFreesurfer@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
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.
On 6/20/17 10:14 AM, marco mcsweeney wrote:
Dear Doug,
thank you so much for getting back to me on this.
I am using Freesurfer version 5.3 with QDEV version 1.4 which gives me both DODS and DOSS design matrix options. I am a little confused about DOSS in QDEC as I have read quite a few publications which have used DOSS in QDEC before. Can you please let me know what the issue with DOSS in QDEC is?
I don't quite remember what was going wrong. It was only under some circumstances, but I can't remember what those were. The correct path forward is to use the mri_glmfit command line program for DOSS.
Also, I have looked at age-cortical thickness interaction using DODS in Qdec with gender and diagnosis as fixed factors, age as covariate (demean age as well) and eTIV as nuisance. I did not find any difference in age-cortical thickness interaction between the 2 groups.
That means you can use DOSS.
Also, I am still a bit unsure about using age as covariate in DOSS through the command line. Is this appropriate given that DOSS assumes the same slope in both groups?
There's nothing you have to do other than specify doss after the FSGD file (and create a new contrast matrix).
Or should age be used as a nuisance factor instead (0 in the design matrix)?
do you mean contrast matrix? if so, then just specify 0 for the nuisance colum(s)
Thank you so much again for your help with this.
Best wishes
Marco
On Tue, Jun 20, 2017 at 2:02 PM, Douglas Greve <greve@nmr.mgh.harvard.edu mailto:greve@nmr.mgh.harvard.edu> wrote:
DOSS should not be used with QDEC as it may or may not do the right thing. Is it not disabled? What version are you using? To answer your question, if you want to use DOSS and you're interested in the diff bet groups, then first test for an interaction between group and covariate. If that is not significant, then use DOSS On 6/20/17 5:05 AM, marco mcsweeney wrote:Dear Freesurfer experts, I have a quick question regarding the appropriate design (DOSS or DODS) to be used with QDEC when comparing 2 groups for cortical thickness and LGI values. So far I have used DODS first to look for an age-cortical thickness interaction in both groups. I used the real ages and also demeaned the age variable in my analysis. I entered age as a covariate and as a nuisance factor also. I have gender and diagnosis as discrete and eTIV as nuisance in both analysis. Neither was significant. After this I used DOSS in QDEC to compare groups groups for both cortical thickness and LGI. My questions relates to the use of age as a covariate in DOSS. Is it appropriate to enter age as covariate in DOSS given that DOSS assumes that both groups have the same slope? I have looked through the archives and online for an answer to this but I am still a bit confused. Any help or advice on this would be greatly appreciated. Thank you so much in advance. Best wishes Marco _______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu <mailto:Freesurfer@nmr.mgh.harvard.edu> https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer <https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer>_______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu <mailto:Freesurfer@nmr.mgh.harvard.edu> https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer <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 <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.
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
Dear Doug,
thank you.
I will go ahead with DOSS through the command line. I did find significant differences (FDR corrected) with DOSS in QDEC. Hopefully they will still be present through the command line. It will be interesting to see if I get the same results.
Sorry but just one more quick question. I think I read on the list that demeaning the age (mean age of all participants - real age) is not required with DOSS. Is this correct? Also, should DODS be used with the demean age covariate rather than the real age when looking for differences in cortical thickness or LGI between 2 groups?
Thank you so much again.
Best wishes
Marco
On Tue, Jun 20, 2017 at 3:29 PM, Douglas Greve greve@nmr.mgh.harvard.edu wrote:
On 6/20/17 10:14 AM, marco mcsweeney wrote:
Dear Doug,
thank you so much for getting back to me on this.
I am using Freesurfer version 5.3 with QDEV version 1.4 which gives me both DODS and DOSS design matrix options. I am a little confused about DOSS in QDEC as I have read quite a few publications which have used DOSS in QDEC before. Can you please let me know what the issue with DOSS in QDEC is?
I don't quite remember what was going wrong. It was only under some circumstances, but I can't remember what those were. The correct path forward is to use the mri_glmfit command line program for DOSS.
Also, I have looked at age-cortical thickness interaction using DODS in Qdec with gender and diagnosis as fixed factors, age as covariate (demean age as well) and eTIV as nuisance. I did not find any difference in age-cortical thickness interaction between the 2 groups.
That means you can use DOSS.
Also, I am still a bit unsure about using age as covariate in DOSS through the command line. Is this appropriate given that DOSS assumes the same slope in both groups?
There's nothing you have to do other than specify doss after the FSGD file (and create a new contrast matrix).
Or should age be used as a nuisance factor instead (0 in the design matrix)?
do you mean contrast matrix? if so, then just specify 0 for the nuisance colum(s)
Thank you so much again for your help with this.
Best wishes
Marco
On Tue, Jun 20, 2017 at 2:02 PM, Douglas Greve greve@nmr.mgh.harvard.edu wrote:
DOSS should not be used with QDEC as it may or may not do the right thing. Is it not disabled? What version are you using?
To answer your question, if you want to use DOSS and you're interested in the diff bet groups, then first test for an interaction between group and covariate. If that is not significant, then use DOSS
On 6/20/17 5:05 AM, marco mcsweeney wrote:
Dear Freesurfer experts,
I have a quick question regarding the appropriate design (DOSS or DODS) to be used with QDEC when comparing 2 groups for cortical thickness and LGI values.
So far I have used DODS first to look for an age-cortical thickness interaction in both groups. I used the real ages and also demeaned the age variable in my analysis. I entered age as a covariate and as a nuisance factor also. I have gender and diagnosis as discrete and eTIV as nuisance in both analysis. Neither was significant.
After this I used DOSS in QDEC to compare groups groups for both cortical thickness and LGI.
My questions relates to the use of age as a covariate in DOSS. Is it appropriate to enter age as covariate in DOSS given that DOSS assumes that both groups have the same slope? I have looked through the archives and online for an answer to this but I am still a bit confused.
Any help or advice on this would be greatly appreciated.
Thank you so much in advance.
Best wishes
Marco
Freesurfer mailing listFreesurfer@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.e du/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.
Freesurfer mailing listFreesurfer@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
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.
On 6/20/17 10:43 AM, marco mcsweeney wrote:
Dear Doug,
thank you.
I will go ahead with DOSS through the command line. I did find significant differences (FDR corrected) with DOSS in QDEC. Hopefully they will still be present through the command line. It will be interesting to see if I get the same results.
Sorry but just one more quick question. I think I read on the list that demeaning the age (mean age of all participants - real age) is not required with DOSS. Is this correct?
Correct.
Also, should DODS be used with the demean age covariate rather than the real age when looking for differences in cortical thickness or LGI between 2 groups?
You should check whether there is an interaction between group and age with DODS (in which case demeaning does not matter). If there is none, then go to DOSS. If there is an interaction, then things get complicated in terms of the interpretation. There is no demeaning or not demeaning that will help this.
Thank you so much again.
Best wishes
Marco
On Tue, Jun 20, 2017 at 3:29 PM, Douglas Greve <greve@nmr.mgh.harvard.edu mailto:greve@nmr.mgh.harvard.edu> wrote:
On 6/20/17 10:14 AM, marco mcsweeney wrote:Dear Doug, thank you so much for getting back to me on this. I am using Freesurfer version 5.3 with QDEV version 1.4 which gives me both DODS and DOSS design matrix options. I am a little confused about DOSS in QDEC as I have read quite a few publications which have used DOSS in QDEC before. Can you please let me know what the issue with DOSS in QDEC is?I don't quite remember what was going wrong. It was only under some circumstances, but I can't remember what those were. The correct path forward is to use the mri_glmfit command line program for DOSS.Also, I have looked at age-cortical thickness interaction using DODS in Qdec with gender and diagnosis as fixed factors, age as covariate (demean age as well) and eTIV as nuisance. I did not find any difference in age-cortical thickness interaction between the 2 groups.That means you can use DOSS.Also, I am still a bit unsure about using age as covariate in DOSS through the command line. Is this appropriate given that DOSS assumes the same slope in both groups?There's nothing you have to do other than specify doss after the FSGD file (and create a new contrast matrix).Or should age be used as a nuisance factor instead (0 in the design matrix)?do you mean contrast matrix? if so, then just specify 0 for the nuisance colum(s)Thank you so much again for your help with this. Best wishes Marco On Tue, Jun 20, 2017 at 2:02 PM, Douglas Greve <greve@nmr.mgh.harvard.edu <mailto:greve@nmr.mgh.harvard.edu>> wrote: DOSS should not be used with QDEC as it may or may not do the right thing. Is it not disabled? What version are you using? To answer your question, if you want to use DOSS and you're interested in the diff bet groups, then first test for an interaction between group and covariate. If that is not significant, then use DOSS On 6/20/17 5:05 AM, marco mcsweeney wrote:Dear Freesurfer experts, I have a quick question regarding the appropriate design (DOSS or DODS) to be used with QDEC when comparing 2 groups for cortical thickness and LGI values. So far I have used DODS first to look for an age-cortical thickness interaction in both groups. I used the real ages and also demeaned the age variable in my analysis. I entered age as a covariate and as a nuisance factor also. I have gender and diagnosis as discrete and eTIV as nuisance in both analysis. Neither was significant. After this I used DOSS in QDEC to compare groups groups for both cortical thickness and LGI. My questions relates to the use of age as a covariate in DOSS. Is it appropriate to enter age as covariate in DOSS given that DOSS assumes that both groups have the same slope? I have looked through the archives and online for an answer to this but I am still a bit confused. Any help or advice on this would be greatly appreciated. Thank you so much in advance. Best wishes Marco _______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu <mailto:Freesurfer@nmr.mgh.harvard.edu> https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer <https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer>_______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu <mailto:Freesurfer@nmr.mgh.harvard.edu> https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer <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 <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. _______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu <mailto:Freesurfer@nmr.mgh.harvard.edu> https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer <https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer>_______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu <mailto:Freesurfer@nmr.mgh.harvard.edu> https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer <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 <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.
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
Dear Doug,
OK thank you so much for all of your help.
Best wishes
Marco
On Tue, Jun 20, 2017 at 3:58 PM, Douglas Greve greve@nmr.mgh.harvard.edu wrote:
On 6/20/17 10:43 AM, marco mcsweeney wrote:
Dear Doug,
thank you.
I will go ahead with DOSS through the command line. I did find significant differences (FDR corrected) with DOSS in QDEC. Hopefully they will still be present through the command line. It will be interesting to see if I get the same results.
Sorry but just one more quick question. I think I read on the list that demeaning the age (mean age of all participants - real age) is not required with DOSS. Is this correct?
Correct.
Also, should DODS be used with the demean age covariate rather than the real age when looking for differences in cortical thickness or LGI between 2 groups?
You should check whether there is an interaction between group and age with DODS (in which case demeaning does not matter). If there is none, then go to DOSS. If there is an interaction, then things get complicated in terms of the interpretation. There is no demeaning or not demeaning that will help this.
Thank you so much again.
Best wishes
Marco
On Tue, Jun 20, 2017 at 3:29 PM, Douglas Greve greve@nmr.mgh.harvard.edu wrote:
On 6/20/17 10:14 AM, marco mcsweeney wrote:
Dear Doug,
thank you so much for getting back to me on this.
I am using Freesurfer version 5.3 with QDEV version 1.4 which gives me both DODS and DOSS design matrix options. I am a little confused about DOSS in QDEC as I have read quite a few publications which have used DOSS in QDEC before. Can you please let me know what the issue with DOSS in QDEC is?
I don't quite remember what was going wrong. It was only under some circumstances, but I can't remember what those were. The correct path forward is to use the mri_glmfit command line program for DOSS.
Also, I have looked at age-cortical thickness interaction using DODS in Qdec with gender and diagnosis as fixed factors, age as covariate (demean age as well) and eTIV as nuisance. I did not find any difference in age-cortical thickness interaction between the 2 groups.
That means you can use DOSS.
Also, I am still a bit unsure about using age as covariate in DOSS through the command line. Is this appropriate given that DOSS assumes the same slope in both groups?
There's nothing you have to do other than specify doss after the FSGD file (and create a new contrast matrix).
Or should age be used as a nuisance factor instead (0 in the design matrix)?
do you mean contrast matrix? if so, then just specify 0 for the nuisance colum(s)
Thank you so much again for your help with this.
Best wishes
Marco
On Tue, Jun 20, 2017 at 2:02 PM, Douglas Greve <greve@nmr.mgh.harvard.edu
wrote:
DOSS should not be used with QDEC as it may or may not do the right thing. Is it not disabled? What version are you using?
To answer your question, if you want to use DOSS and you're interested in the diff bet groups, then first test for an interaction between group and covariate. If that is not significant, then use DOSS
On 6/20/17 5:05 AM, marco mcsweeney wrote:
Dear Freesurfer experts,
I have a quick question regarding the appropriate design (DOSS or DODS) to be used with QDEC when comparing 2 groups for cortical thickness and LGI values.
So far I have used DODS first to look for an age-cortical thickness interaction in both groups. I used the real ages and also demeaned the age variable in my analysis. I entered age as a covariate and as a nuisance factor also. I have gender and diagnosis as discrete and eTIV as nuisance in both analysis. Neither was significant.
After this I used DOSS in QDEC to compare groups groups for both cortical thickness and LGI.
My questions relates to the use of age as a covariate in DOSS. Is it appropriate to enter age as covariate in DOSS given that DOSS assumes that both groups have the same slope? I have looked through the archives and online for an answer to this but I am still a bit confused.
Any help or advice on this would be greatly appreciated.
Thank you so much in advance.
Best wishes
Marco
Freesurfer mailing listFreesurfer@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.e du/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.
Freesurfer mailing listFreesurfer@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.e du/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.
Freesurfer mailing listFreesurfer@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
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.
Dear Doug,
I have a few final questions, my apologies.
The study I am working on is a retrospective study, which is not ideal. As a result, although all of the individuals in the study were scanned on the same 3T scanner, all of the healthy controls have 1.6mm slice thickness and most of the patients have 1mm and a few have 800.00 µm slice thickness. All subjects (N = 58, patients = 29) have FOV = 256, voxel sizes 1.000, 1.000, 1.000.
Should I control for slice thickness in the analysis using the appropriate contrast matrix?
Will this have a significant impact on the GLM results?
Also, is it possible to check slice thickness for each scan using the mri_info flag or other type of flag? I have tried using mri_info to double check slice thickness for each scan but it doesn't seem to give this in the output.
Also, would you advise using eTIV or Total gray matter volume as a nuisance factor in the analysis.
Thank you so much again for your time.
Best wishes
Marco
On Tue, Jun 20, 2017 at 3:58 PM, Douglas Greve greve@nmr.mgh.harvard.edu wrote:
On 6/20/17 10:43 AM, marco mcsweeney wrote:
Dear Doug,
thank you.
I will go ahead with DOSS through the command line. I did find significant differences (FDR corrected) with DOSS in QDEC. Hopefully they will still be present through the command line. It will be interesting to see if I get the same results.
Sorry but just one more quick question. I think I read on the list that demeaning the age (mean age of all participants - real age) is not required with DOSS. Is this correct?
Correct.
Also, should DODS be used with the demean age covariate rather than the real age when looking for differences in cortical thickness or LGI between 2 groups?
You should check whether there is an interaction between group and age with DODS (in which case demeaning does not matter). If there is none, then go to DOSS. If there is an interaction, then things get complicated in terms of the interpretation. There is no demeaning or not demeaning that will help this.
Thank you so much again.
Best wishes
Marco
On Tue, Jun 20, 2017 at 3:29 PM, Douglas Greve greve@nmr.mgh.harvard.edu wrote:
On 6/20/17 10:14 AM, marco mcsweeney wrote:
Dear Doug,
thank you so much for getting back to me on this.
I am using Freesurfer version 5.3 with QDEV version 1.4 which gives me both DODS and DOSS design matrix options. I am a little confused about DOSS in QDEC as I have read quite a few publications which have used DOSS in QDEC before. Can you please let me know what the issue with DOSS in QDEC is?
I don't quite remember what was going wrong. It was only under some circumstances, but I can't remember what those were. The correct path forward is to use the mri_glmfit command line program for DOSS.
Also, I have looked at age-cortical thickness interaction using DODS in Qdec with gender and diagnosis as fixed factors, age as covariate (demean age as well) and eTIV as nuisance. I did not find any difference in age-cortical thickness interaction between the 2 groups.
That means you can use DOSS.
Also, I am still a bit unsure about using age as covariate in DOSS through the command line. Is this appropriate given that DOSS assumes the same slope in both groups?
There's nothing you have to do other than specify doss after the FSGD file (and create a new contrast matrix).
Or should age be used as a nuisance factor instead (0 in the design matrix)?
do you mean contrast matrix? if so, then just specify 0 for the nuisance colum(s)
Thank you so much again for your help with this.
Best wishes
Marco
On Tue, Jun 20, 2017 at 2:02 PM, Douglas Greve <greve@nmr.mgh.harvard.edu
wrote:
DOSS should not be used with QDEC as it may or may not do the right thing. Is it not disabled? What version are you using?
To answer your question, if you want to use DOSS and you're interested in the diff bet groups, then first test for an interaction between group and covariate. If that is not significant, then use DOSS
On 6/20/17 5:05 AM, marco mcsweeney wrote:
Dear Freesurfer experts,
I have a quick question regarding the appropriate design (DOSS or DODS) to be used with QDEC when comparing 2 groups for cortical thickness and LGI values.
So far I have used DODS first to look for an age-cortical thickness interaction in both groups. I used the real ages and also demeaned the age variable in my analysis. I entered age as a covariate and as a nuisance factor also. I have gender and diagnosis as discrete and eTIV as nuisance in both analysis. Neither was significant.
After this I used DOSS in QDEC to compare groups groups for both cortical thickness and LGI.
My questions relates to the use of age as a covariate in DOSS. Is it appropriate to enter age as covariate in DOSS given that DOSS assumes that both groups have the same slope? I have looked through the archives and online for an answer to this but I am still a bit confused.
Any help or advice on this would be greatly appreciated.
Thank you so much in advance.
Best wishes
Marco
Freesurfer mailing listFreesurfer@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.e du/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.
Freesurfer mailing listFreesurfer@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.e du/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.
Freesurfer mailing listFreesurfer@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
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.
mri_info --sres file.mgz will give you the slice thickness
you should at least control for the slice thickness by creating a new binary factor, so if you had 2 groups, you'd have 4 groups (this is the design matrix not the contrast matrix)
On 6/20/17 11:58 AM, marco mcsweeney wrote:
Dear Doug,
I have a few final questions, my apologies.
The study I am working on is a retrospective study, which is not ideal. As a result, although all of the individuals in the study were scanned on the same 3T scanner, all of the healthy controls have 1.6mm slice thickness and most of the patients have 1mm and a few have 800.00 µm slice thickness. All subjects (N = 58, patients = 29) have FOV = 256, voxel sizes 1.000, 1.000, 1.000.
Should I control for slice thickness in the analysis using the appropriate contrast matrix?
Will this have a significant impact on the GLM results?
Also, is it possible to check slice thickness for each scan using the mri_info flag or other type of flag? I have tried using mri_info to double check slice thickness for each scan but it doesn't seem to give this in the output.
Also, would you advise using eTIV or Total gray matter volume as a nuisance factor in the analysis.
Thank you so much again for your time.
Best wishes
Marco
On Tue, Jun 20, 2017 at 3:58 PM, Douglas Greve <greve@nmr.mgh.harvard.edu mailto:greve@nmr.mgh.harvard.edu> wrote:
On 6/20/17 10:43 AM, marco mcsweeney wrote:Dear Doug, thank you. I will go ahead with DOSS through the command line. I did find significant differences (FDR corrected) with DOSS in QDEC. Hopefully they will still be present through the command line. It will be interesting to see if I get the same results. Sorry but just one more quick question. I think I read on the list that demeaning the age (mean age of all participants - real age) is not required with DOSS. Is this correct?Correct.Also, should DODS be used with the demean age covariate rather than the real age when looking for differences in cortical thickness or LGI between 2 groups?You should check whether there is an interaction between group and age with DODS (in which case demeaning does not matter). If there is none, then go to DOSS. If there is an interaction, then things get complicated in terms of the interpretation. There is no demeaning or not demeaning that will help this.Thank you so much again. Best wishes Marco On Tue, Jun 20, 2017 at 3:29 PM, Douglas Greve <greve@nmr.mgh.harvard.edu <mailto:greve@nmr.mgh.harvard.edu>> wrote: On 6/20/17 10:14 AM, marco mcsweeney wrote:Dear Doug, thank you so much for getting back to me on this. I am using Freesurfer version 5.3 with QDEV version 1.4 which gives me both DODS and DOSS design matrix options. I am a little confused about DOSS in QDEC as I have read quite a few publications which have used DOSS in QDEC before. Can you please let me know what the issue with DOSS in QDEC is?I don't quite remember what was going wrong. It was only under some circumstances, but I can't remember what those were. The correct path forward is to use the mri_glmfit command line program for DOSS.Also, I have looked at age-cortical thickness interaction using DODS in Qdec with gender and diagnosis as fixed factors, age as covariate (demean age as well) and eTIV as nuisance. I did not find any difference in age-cortical thickness interaction between the 2 groups.That means you can use DOSS.Also, I am still a bit unsure about using age as covariate in DOSS through the command line. Is this appropriate given that DOSS assumes the same slope in both groups?There's nothing you have to do other than specify doss after the FSGD file (and create a new contrast matrix).Or should age be used as a nuisance factor instead (0 in the design matrix)?do you mean contrast matrix? if so, then just specify 0 for the nuisance colum(s)Thank you so much again for your help with this. Best wishes Marco On Tue, Jun 20, 2017 at 2:02 PM, Douglas Greve <greve@nmr.mgh.harvard.edu <mailto:greve@nmr.mgh.harvard.edu>> wrote: DOSS should not be used with QDEC as it may or may not do the right thing. Is it not disabled? What version are you using? To answer your question, if you want to use DOSS and you're interested in the diff bet groups, then first test for an interaction between group and covariate. If that is not significant, then use DOSS On 6/20/17 5:05 AM, marco mcsweeney wrote:Dear Freesurfer experts, I have a quick question regarding the appropriate design (DOSS or DODS) to be used with QDEC when comparing 2 groups for cortical thickness and LGI values. So far I have used DODS first to look for an age-cortical thickness interaction in both groups. I used the real ages and also demeaned the age variable in my analysis. I entered age as a covariate and as a nuisance factor also. I have gender and diagnosis as discrete and eTIV as nuisance in both analysis. Neither was significant. After this I used DOSS in QDEC to compare groups groups for both cortical thickness and LGI. My questions relates to the use of age as a covariate in DOSS. Is it appropriate to enter age as covariate in DOSS given that DOSS assumes that both groups have the same slope? I have looked through the archives and online for an answer to this but I am still a bit confused. Any help or advice on this would be greatly appreciated. Thank you so much in advance. Best wishes Marco _______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu <mailto:Freesurfer@nmr.mgh.harvard.edu> https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer <https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer>_______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu <mailto:Freesurfer@nmr.mgh.harvard.edu> https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer <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 <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. _______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu <mailto:Freesurfer@nmr.mgh.harvard.edu> https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer <https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer>_______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu <mailto:Freesurfer@nmr.mgh.harvard.edu> https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer <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 <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. _______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu <mailto:Freesurfer@nmr.mgh.harvard.edu> https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer <https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer>_______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu <mailto:Freesurfer@nmr.mgh.harvard.edu> https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer <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 <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.
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
Dear Doug,
thanks again. I used mri_info --sres orig.mgz (T1.mgz) and got output 1. I assume this is 1mm thickness?
Also, in terms of using eTIV or total gray matter volume, which do you think would be the best to use as nuisance factor when looking at cortical thickness as well as LGI?
Thanks again
Marco
On Tue, Jun 20, 2017 at 7:15 PM, Douglas Greve greve@nmr.mgh.harvard.edu wrote:
mri_info --sres file.mgz will give you the slice thickness
you should at least control for the slice thickness by creating a new binary factor, so if you had 2 groups, you'd have 4 groups (this is the design matrix not the contrast matrix)
On 6/20/17 11:58 AM, marco mcsweeney wrote:
Dear Doug,
I have a few final questions, my apologies.
The study I am working on is a retrospective study, which is not ideal. As a result, although all of the individuals in the study were scanned on the same 3T scanner, all of the healthy controls have 1.6mm slice thickness and most of the patients have 1mm and a few have 800.00 µm slice thickness. All subjects (N = 58, patients = 29) have FOV = 256, voxel sizes 1.000, 1.000, 1.000.
Should I control for slice thickness in the analysis using the appropriate contrast matrix?
Will this have a significant impact on the GLM results?
Also, is it possible to check slice thickness for each scan using the mri_info flag or other type of flag? I have tried using mri_info to double check slice thickness for each scan but it doesn't seem to give this in the output.
Also, would you advise using eTIV or Total gray matter volume as a nuisance factor in the analysis.
Thank you so much again for your time.
Best wishes
Marco
On Tue, Jun 20, 2017 at 3:58 PM, Douglas Greve greve@nmr.mgh.harvard.edu wrote:
On 6/20/17 10:43 AM, marco mcsweeney wrote:
Dear Doug,
thank you.
I will go ahead with DOSS through the command line. I did find significant differences (FDR corrected) with DOSS in QDEC. Hopefully they will still be present through the command line. It will be interesting to see if I get the same results.
Sorry but just one more quick question. I think I read on the list that demeaning the age (mean age of all participants - real age) is not required with DOSS. Is this correct?
Correct.
Also, should DODS be used with the demean age covariate rather than the real age when looking for differences in cortical thickness or LGI between 2 groups?
You should check whether there is an interaction between group and age with DODS (in which case demeaning does not matter). If there is none, then go to DOSS. If there is an interaction, then things get complicated in terms of the interpretation. There is no demeaning or not demeaning that will help this.
Thank you so much again.
Best wishes
Marco
On Tue, Jun 20, 2017 at 3:29 PM, Douglas Greve <greve@nmr.mgh.harvard.edu
wrote:
On 6/20/17 10:14 AM, marco mcsweeney wrote:
Dear Doug,
thank you so much for getting back to me on this.
I am using Freesurfer version 5.3 with QDEV version 1.4 which gives me both DODS and DOSS design matrix options. I am a little confused about DOSS in QDEC as I have read quite a few publications which have used DOSS in QDEC before. Can you please let me know what the issue with DOSS in QDEC is?
I don't quite remember what was going wrong. It was only under some circumstances, but I can't remember what those were. The correct path forward is to use the mri_glmfit command line program for DOSS.
Also, I have looked at age-cortical thickness interaction using DODS in Qdec with gender and diagnosis as fixed factors, age as covariate (demean age as well) and eTIV as nuisance. I did not find any difference in age-cortical thickness interaction between the 2 groups.
That means you can use DOSS.
Also, I am still a bit unsure about using age as covariate in DOSS through the command line. Is this appropriate given that DOSS assumes the same slope in both groups?
There's nothing you have to do other than specify doss after the FSGD file (and create a new contrast matrix).
Or should age be used as a nuisance factor instead (0 in the design matrix)?
do you mean contrast matrix? if so, then just specify 0 for the nuisance colum(s)
Thank you so much again for your help with this.
Best wishes
Marco
On Tue, Jun 20, 2017 at 2:02 PM, Douglas Greve < greve@nmr.mgh.harvard.edu> wrote:
DOSS should not be used with QDEC as it may or may not do the right thing. Is it not disabled? What version are you using?
To answer your question, if you want to use DOSS and you're interested in the diff bet groups, then first test for an interaction between group and covariate. If that is not significant, then use DOSS
On 6/20/17 5:05 AM, marco mcsweeney wrote:
Dear Freesurfer experts,
I have a quick question regarding the appropriate design (DOSS or DODS) to be used with QDEC when comparing 2 groups for cortical thickness and LGI values.
So far I have used DODS first to look for an age-cortical thickness interaction in both groups. I used the real ages and also demeaned the age variable in my analysis. I entered age as a covariate and as a nuisance factor also. I have gender and diagnosis as discrete and eTIV as nuisance in both analysis. Neither was significant.
After this I used DOSS in QDEC to compare groups groups for both cortical thickness and LGI.
My questions relates to the use of age as a covariate in DOSS. Is it appropriate to enter age as covariate in DOSS given that DOSS assumes that both groups have the same slope? I have looked through the archives and online for an answer to this but I am still a bit confused.
Any help or advice on this would be greatly appreciated.
Thank you so much in advance.
Best wishes
Marco
Freesurfer mailing listFreesurfer@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.e du/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.
Freesurfer mailing listFreesurfer@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.e du/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.
Freesurfer mailing listFreesurfer@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.e du/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.
Freesurfer mailing listFreesurfer@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
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.
On 6/20/17 2:42 PM, marco mcsweeney wrote:
Dear Doug,
thanks again. I used mri_info --sres orig.mgz (T1.mgz) and got output 1. I assume this is 1mm thickness?
yes
Also, in terms of using eTIV or total gray matter volume, which do you think would be the best to use as nuisance factor when looking at cortical thickness as well as LGI?
for thickness, do not use either since thickness does not change with head or brain size. I don't have any intuition as to which would be better for LGI.
Thanks again
Marco
On Tue, Jun 20, 2017 at 7:15 PM, Douglas Greve <greve@nmr.mgh.harvard.edu mailto:greve@nmr.mgh.harvard.edu> wrote:
mri_info --sres file.mgz will give you the slice thickness you should at least control for the slice thickness by creating a new binary factor, so if you had 2 groups, you'd have 4 groups (this is the design matrix not the contrast matrix) On 6/20/17 11:58 AM, marco mcsweeney wrote:Dear Doug, I have a few final questions, my apologies. The study I am working on is a retrospective study, which is not ideal. As a result, although all of the individuals in the study were scanned on the same 3T scanner, all of the healthy controls have 1.6mm slice thickness and most of the patients have 1mm and a few have 800.00 µm slice thickness. All subjects (N = 58, patients = 29) have FOV = 256, voxel sizes 1.000, 1.000, 1.000. Should I control for slice thickness in the analysis using the appropriate contrast matrix? Will this have a significant impact on the GLM results? Also, is it possible to check slice thickness for each scan using the mri_info flag or other type of flag? I have tried using mri_info to double check slice thickness for each scan but it doesn't seem to give this in the output. Also, would you advise using eTIV or Total gray matter volume as a nuisance factor in the analysis. Thank you so much again for your time. Best wishes Marco On Tue, Jun 20, 2017 at 3:58 PM, Douglas Greve <greve@nmr.mgh.harvard.edu <mailto:greve@nmr.mgh.harvard.edu>> wrote: On 6/20/17 10:43 AM, marco mcsweeney wrote:Dear Doug, thank you. I will go ahead with DOSS through the command line. I did find significant differences (FDR corrected) with DOSS in QDEC. Hopefully they will still be present through the command line. It will be interesting to see if I get the same results. Sorry but just one more quick question. I think I read on the list that demeaning the age (mean age of all participants - real age) is not required with DOSS. Is this correct?Correct.Also, should DODS be used with the demean age covariate rather than the real age when looking for differences in cortical thickness or LGI between 2 groups?You should check whether there is an interaction between group and age with DODS (in which case demeaning does not matter). If there is none, then go to DOSS. If there is an interaction, then things get complicated in terms of the interpretation. There is no demeaning or not demeaning that will help this.Thank you so much again. Best wishes Marco On Tue, Jun 20, 2017 at 3:29 PM, Douglas Greve <greve@nmr.mgh.harvard.edu <mailto:greve@nmr.mgh.harvard.edu>> wrote: On 6/20/17 10:14 AM, marco mcsweeney wrote:Dear Doug, thank you so much for getting back to me on this. I am using Freesurfer version 5.3 with QDEV version 1.4 which gives me both DODS and DOSS design matrix options. I am a little confused about DOSS in QDEC as I have read quite a few publications which have used DOSS in QDEC before. Can you please let me know what the issue with DOSS in QDEC is?I don't quite remember what was going wrong. It was only under some circumstances, but I can't remember what those were. The correct path forward is to use the mri_glmfit command line program for DOSS.Also, I have looked at age-cortical thickness interaction using DODS in Qdec with gender and diagnosis as fixed factors, age as covariate (demean age as well) and eTIV as nuisance. I did not find any difference in age-cortical thickness interaction between the 2 groups.That means you can use DOSS.Also, I am still a bit unsure about using age as covariate in DOSS through the command line. Is this appropriate given that DOSS assumes the same slope in both groups?There's nothing you have to do other than specify doss after the FSGD file (and create a new contrast matrix).Or should age be used as a nuisance factor instead (0 in the design matrix)?do you mean contrast matrix? if so, then just specify 0 for the nuisance colum(s)Thank you so much again for your help with this. Best wishes Marco On Tue, Jun 20, 2017 at 2:02 PM, Douglas Greve <greve@nmr.mgh.harvard.edu <mailto:greve@nmr.mgh.harvard.edu>> wrote: DOSS should not be used with QDEC as it may or may not do the right thing. Is it not disabled? What version are you using? To answer your question, if you want to use DOSS and you're interested in the diff bet groups, then first test for an interaction between group and covariate. If that is not significant, then use DOSS On 6/20/17 5:05 AM, marco mcsweeney wrote:Dear Freesurfer experts, I have a quick question regarding the appropriate design (DOSS or DODS) to be used with QDEC when comparing 2 groups for cortical thickness and LGI values. So far I have used DODS first to look for an age-cortical thickness interaction in both groups. I used the real ages and also demeaned the age variable in my analysis. I entered age as a covariate and as a nuisance factor also. I have gender and diagnosis as discrete and eTIV as nuisance in both analysis. Neither was significant. After this I used DOSS in QDEC to compare groups groups for both cortical thickness and LGI. My questions relates to the use of age as a covariate in DOSS. Is it appropriate to enter age as covariate in DOSS given that DOSS assumes that both groups have the same slope? I have looked through the archives and online for an answer to this but I am still a bit confused. Any help or advice on this would be greatly appreciated. Thank you so much in advance. Best wishes Marco _______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu <mailto:Freesurfer@nmr.mgh.harvard.edu> https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer <https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer>_______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu <mailto:Freesurfer@nmr.mgh.harvard.edu> https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer <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 <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. _______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu <mailto:Freesurfer@nmr.mgh.harvard.edu> https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer <https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer>_______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu <mailto:Freesurfer@nmr.mgh.harvard.edu> https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer <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 <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. _______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu <mailto:Freesurfer@nmr.mgh.harvard.edu> https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer <https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer>_______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu <mailto:Freesurfer@nmr.mgh.harvard.edu> https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer <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 <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. _______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu <mailto:Freesurfer@nmr.mgh.harvard.edu> https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer <https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer>_______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu <mailto:Freesurfer@nmr.mgh.harvard.edu> https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer <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 <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.
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
Dear Doug,
thanks again.
Will look into which is the best for LGI (if any).
Best wishes
Marco
On Tue, Jun 20, 2017 at 7:48 PM, Douglas Greve greve@nmr.mgh.harvard.edu wrote:
On 6/20/17 2:42 PM, marco mcsweeney wrote:
Dear Doug,
thanks again. I used mri_info --sres orig.mgz (T1.mgz) and got output 1. I assume this is 1mm thickness?
yes
Also, in terms of using eTIV or total gray matter volume, which do you think would be the best to use as nuisance factor when looking at cortical thickness as well as LGI?
for thickness, do not use either since thickness does not change with head or brain size. I don't have any intuition as to which would be better for LGI.
Thanks again
Marco
On Tue, Jun 20, 2017 at 7:15 PM, Douglas Greve greve@nmr.mgh.harvard.edu wrote:
mri_info --sres file.mgz will give you the slice thickness
you should at least control for the slice thickness by creating a new binary factor, so if you had 2 groups, you'd have 4 groups (this is the design matrix not the contrast matrix)
On 6/20/17 11:58 AM, marco mcsweeney wrote:
Dear Doug,
I have a few final questions, my apologies.
The study I am working on is a retrospective study, which is not ideal. As a result, although all of the individuals in the study were scanned on the same 3T scanner, all of the healthy controls have 1.6mm slice thickness and most of the patients have 1mm and a few have 800.00 µm slice thickness. All subjects (N = 58, patients = 29) have FOV = 256, voxel sizes 1.000, 1.000, 1.000.
Should I control for slice thickness in the analysis using the appropriate contrast matrix?
Will this have a significant impact on the GLM results?
Also, is it possible to check slice thickness for each scan using the mri_info flag or other type of flag? I have tried using mri_info to double check slice thickness for each scan but it doesn't seem to give this in the output.
Also, would you advise using eTIV or Total gray matter volume as a nuisance factor in the analysis.
Thank you so much again for your time.
Best wishes
Marco
On Tue, Jun 20, 2017 at 3:58 PM, Douglas Greve <greve@nmr.mgh.harvard.edu
wrote:
On 6/20/17 10:43 AM, marco mcsweeney wrote:
Dear Doug,
thank you.
I will go ahead with DOSS through the command line. I did find significant differences (FDR corrected) with DOSS in QDEC. Hopefully they will still be present through the command line. It will be interesting to see if I get the same results.
Sorry but just one more quick question. I think I read on the list that demeaning the age (mean age of all participants - real age) is not required with DOSS. Is this correct?
Correct.
Also, should DODS be used with the demean age covariate rather than the real age when looking for differences in cortical thickness or LGI between 2 groups?
You should check whether there is an interaction between group and age with DODS (in which case demeaning does not matter). If there is none, then go to DOSS. If there is an interaction, then things get complicated in terms of the interpretation. There is no demeaning or not demeaning that will help this.
Thank you so much again.
Best wishes
Marco
On Tue, Jun 20, 2017 at 3:29 PM, Douglas Greve < greve@nmr.mgh.harvard.edu> wrote:
On 6/20/17 10:14 AM, marco mcsweeney wrote:
Dear Doug,
thank you so much for getting back to me on this.
I am using Freesurfer version 5.3 with QDEV version 1.4 which gives me both DODS and DOSS design matrix options. I am a little confused about DOSS in QDEC as I have read quite a few publications which have used DOSS in QDEC before. Can you please let me know what the issue with DOSS in QDEC is?
I don't quite remember what was going wrong. It was only under some circumstances, but I can't remember what those were. The correct path forward is to use the mri_glmfit command line program for DOSS.
Also, I have looked at age-cortical thickness interaction using DODS in Qdec with gender and diagnosis as fixed factors, age as covariate (demean age as well) and eTIV as nuisance. I did not find any difference in age-cortical thickness interaction between the 2 groups.
That means you can use DOSS.
Also, I am still a bit unsure about using age as covariate in DOSS through the command line. Is this appropriate given that DOSS assumes the same slope in both groups?
There's nothing you have to do other than specify doss after the FSGD file (and create a new contrast matrix).
Or should age be used as a nuisance factor instead (0 in the design matrix)?
do you mean contrast matrix? if so, then just specify 0 for the nuisance colum(s)
Thank you so much again for your help with this.
Best wishes
Marco
On Tue, Jun 20, 2017 at 2:02 PM, Douglas Greve < greve@nmr.mgh.harvard.edu> wrote:
DOSS should not be used with QDEC as it may or may not do the right thing. Is it not disabled? What version are you using?
To answer your question, if you want to use DOSS and you're interested in the diff bet groups, then first test for an interaction between group and covariate. If that is not significant, then use DOSS
On 6/20/17 5:05 AM, marco mcsweeney wrote:
Dear Freesurfer experts,
I have a quick question regarding the appropriate design (DOSS or DODS) to be used with QDEC when comparing 2 groups for cortical thickness and LGI values.
So far I have used DODS first to look for an age-cortical thickness interaction in both groups. I used the real ages and also demeaned the age variable in my analysis. I entered age as a covariate and as a nuisance factor also. I have gender and diagnosis as discrete and eTIV as nuisance in both analysis. Neither was significant.
After this I used DOSS in QDEC to compare groups groups for both cortical thickness and LGI.
My questions relates to the use of age as a covariate in DOSS. Is it appropriate to enter age as covariate in DOSS given that DOSS assumes that both groups have the same slope? I have looked through the archives and online for an answer to this but I am still a bit confused.
Any help or advice on this would be greatly appreciated.
Thank you so much in advance.
Best wishes
Marco
Freesurfer mailing listFreesurfer@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.e du/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.
Freesurfer mailing listFreesurfer@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.e du/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.
Freesurfer mailing listFreesurfer@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.e du/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.
Freesurfer mailing listFreesurfer@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.e du/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.
Freesurfer mailing listFreesurfer@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
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.
freesurfer@nmr.mgh.harvard.edu
-
Douglas Greve -
marco mcsweeney