Re: [Freesurfer] repeated measures ANOVA 2x2 two groups two times {Disarmed}

[Agurne Sampedro Calvete]

External Email - Use Caution

I found it here:
https://secure-web.cisco.com/1Pobz-1CRgeaN25ufKu4UcOwbvjX2LjQiAacXtHLnJRq3FQhxMBarP9V62pvmTN-oTfoNLhHO4A5InMyZX-CjChowd6t4LnXQVq71s1pWtfaI0sRBlnMEein6reZM__dj0bhTOz-ln6_F607244YC4DVDD0Nco3HESUEIMLe4SvbfGSa-LaRnXl2QLem4nS4iHic8qO4tkACjlueIIBYKzg/https%3A%2F%2Fsurfer.nmr.mgh.harvard.edu%2Ffswiki%2FFsTutorial%2FGroupAnalysis

I may have misunderstood.

Thank you very much.

Regards,

Agurne





Neuropsychology of Severe Medical Conditions Research Team

Dpt of Methods and Experimental Psychology
Faculty of Psychology and Education

University of Deusto

neurolab.deusto.es



Avda. de las Universidades 24,

48007, Bilbao (Spain)

Phone: +34 944139000 (ext. 3015)


El vie., 6 nov. 2020 a las 15:57, Douglas N. Greve ()
escribió:

> Oh, that's an error, you do not have to include --cache-in. Where is that
> page?
>
> On 11/5/2020 10:32 AM, Agurne Sampedro Calvete wrote:
>
> External Email - Use Caution
> Dear Douglas,
>
> Thank you very much for your rapid response. I have another question.
>
> I am following the FS Repeated Measures tutorial and therefore, I ran
> this: mris_preproc --target fsaverage --hemi lh --meas thickness --out
> lh.thickness.mgh --fsgd rmanova.fsgd.
> However, in the  FS Group Analysis tutorial, it says that you must include
> "--cache-in thickness.fwhm10.fsaverage" if the data was preprocessed with
> recon-all with the -qcache option, which was my case. As I am doing
> longitudinal analyses, do I have to skip this  option and just follow the
> Repeated Measures tutorial or do I have to include the --cache option?
>
> Thank you very much.
> Best regards,
>
> Agurne
>
>
>
>
>
> Neuropsychology of Severe Medical Conditions Research Team
>
> Dpt of Methods and Experimental Psychology
> Faculty of Psychology and Education
>
> University of Deusto
>
> neurolab.deusto.es
> 
>
>
>
> Avda. de las Universidades 24,
>
> 48007, Bilbao (Spain)
>
> Phone: +34 944139000 (ext. 3015)
>
>
> El jue., 5 nov. 2020 a las 15:58, Douglas N. Greve (<
> dgr...@mgh.harvard.edu>) escribió:
>
>> In this case, the paired-t is identical to RM 2x2 ANOVA. But to answer
>> your question, the first example you give is the right way to handle it.
>>
>>
>> On 11/5/2020 4:32 AM, Agurne Sampedro Calvete wrote:
>>
>> External Email - Use Caution
>> Dear Freesurfer experts,
>>
>> I am trying to do a longitudinal repeated measures ANOVA 2x2, with two
>> groups (experimental vs control) and two time points (pre and post
>> treatment).
>>
>> Although I have run a paired t-test, now I would like to do a repeated
>> measures anova 2x2, since I am doing the same analyses with other data in
>> other softwares.
>>
>> I am trying to follow the tutorial *MailScanner has detected a possible
>> fraud attempt from "secure-web.cisco.com" claiming to be*
>> https://secure-web.cisco.com/1wVu5Gz4iksiDQpw3WC4tPugyfo6UlKc1Cl2Uk8w-z5ecyacsquyV0qU-68vnyygnGZ5iW0U3JiCT8cuT-0PYFtCY0uM9QqGWNk0fyzvLK5iwxb3F3qpx1hrVR1EVrTPlulNOK3QZPF_uaTdx_pd9qcWxIKySbSIJ7Mysranzw_KddhPOlGqThxPB6Vhk3zUUVSBc6n0LOLFxdFB58nCLbw/https%3A%2F%2Fsurfer.nmr.mgh.harvard.edu%2Ffswiki%2FRepeatedMeasuresAnova
>> ,
>> but I have several doubts when designing the FSGD file as well as the
>> contrasts.
>>
>> How should it be the FSGD file?
>>
>> I have seen in a previous post the following design for the same number
>> fo groups and time points:
>>
>> GroupDescriptorFile 1
>>
>> Class Subject1
>>
>> Class Subject2
>>
>> Variables Tp1vsTp2.GrExp
>>Tp1vsTp2.GrControl
>>
>> Input   Subject1_time1  Subject1
>> 1 0
>>
>> Input   Subject1_time2  Subject1-1
>> 0
>>
>> Input   Subject2_time1  Subject2
>> 0 1
>>
>> Input   Subject2_time2  Subject02   0
>> -1
>>
>>
>>
>> Design matrix:
>>
>> 1 0 1 0
>>
>> 1 0 -1 0
>>
>> 0 1 0 1
>>
>> 0 1 0 -1
>>
>> Contrasts (.mtx files):
>> Tp1vsTp2.GrExp0 0 1 0
>> Tp1vsTp2.GrControl  0 0 0 -1
>> Interaction  0 0 1 -1
>>
>> Would this design be the correct one to use in this case?
>>
>> Alternatively, could it be this the correct one?
>>
>> GroupDescriptorFile 1
>>
>> Class Subject1
>>
>> Class Subject2
>>
>> Variables Tp1vsTp2
>>  Groupxtime
>>
>> Input   Subject1_time1  Subject1 1

Re: [Freesurfer] repeated measures ANOVA 2x2 two groups two times {Disarmed}

[Douglas N. Greve]

Oh, that's an error, you do not have to include --cache-in. Where is 
that page?


On 11/5/2020 10:32 AM, Agurne Sampedro Calvete wrote:


External Email - Use Caution

Dear Douglas,

Thank you very much for your rapid response. I have another question.

I am following the FS Repeated Measures tutorial and therefore, I ran 
this: mris_preproc --target fsaverage --hemi lh --meas thickness --out 
lh.thickness.mgh --fsgd rmanova.fsgd.
However, in the FS Group Analysis tutorial, it says that you must 
include "--cache-in thickness.fwhm10.fsaverage" if the data was 
preprocessed with recon-all with the -qcache option, which was my 
case. As I am doing longitudinal analyses, do I have to skip this 
option and just follow the Repeated Measures tutorial or do I have to 
include the --cache option?


Thank you very much.
Best regards,

Agurne

Neuropsychology of Severe Medical Conditions Research Team

Dpt of Methods and Experimental Psychology
Faculty of Psychology and Education

University of Deusto

neurolab.deusto.es 



Avda. de las Universidades 24,

48007, Bilbao (Spain)

Phone: +34 944139000 (ext. 3015)



El jue., 5 nov. 2020 a las 15:58, Douglas N. Greve 
(mailto:dgr...@mgh.harvard.edu>>) escribió:


In this case, the paired-t is identical to RM 2x2 ANOVA. But to
answer your question, the first example you give is the right way
to handle it.


On 11/5/2020 4:32 AM, Agurne Sampedro Calvete wrote:


External Email - Use Caution

Dear Freesurfer experts,

I am trying to do a longitudinal repeated measures ANOVA 2x2,
with two groups (experimental vs control) and two time points
(pre and post treatment).

Although I have run a paired t-test, now I would like to do a
repeated measures anova 2x2, since I am doing the same analyses
with other data in other softwares.

I am trying to follow the tutorial *MailScanner has detected a
possible fraud attempt from "secure-web.cisco.com" claiming to
be*
https://surfer.nmr.mgh.harvard.edu/fswiki/RepeatedMeasuresAnova

,
but I have several doubts when designing the FSGD file as well as
the contrasts.

How should it be the FSGD file?

I have seen in a previous post the following design for the same
number fo groups and time points:

GroupDescriptorFile 1

Class Subject1

Class Subject2

Variables
Tp1vsTp2.GrExp
   Tp1vsTp2.GrControl

Input   Subject1_time1  Subject1
1 0

Input   Subject1_time2  Subject1 -1    0

Input   Subject2_time1  Subject2 0 1

Input   Subject2_time2  Subject02
 0 -1

Design matrix:

1 0 1 0

1 0 -1 0

0 1 0 1

0 1 0 -1


Contrasts (.mtx files):
Tp1vsTp2.GrExp   0 0 1 0
Tp1vsTp2.GrControl 0 0 0 -1
Interaction         0 0 1 -1

Would this design be the correct one to use in this case?

Alternatively, could it be this the correct one?

GroupDescriptorFile 1

Class Subject1

Class Subject2

Variables Tp1vsTp2
 Groupxtime

Input  Subject1_time1  Subject1
1    1

Input  Subject1_time2  Subject1 -1 *-1*

Input   Subject2_time1  Subject2   1 *-1*

Input  Subject2_time2  Subject02  -1  1

Design matrix:

1 0  1 1

1 0 -1 -1

0 1  1 -1

0 1  -1 1

Contrasts (.mtx files):

Tp1vsTp2         0 0 1 0

Groupxtime Interaction   0 0 0 1


Thank you very much in advance.

Kind regards,

Agurne

Neuropsychology of Severe Medical Conditions Research Team

Dpt of Methods and Experimental Psychology
Faculty of Psychology and Education

University of Deusto

neurolab.deusto.es



Avda. de las Universidades 24,

48007, Bilbao (Spain)

Phone: +34 944139000 (ext. 3015)


___
   

Re: [Freesurfer] repeated measures ANOVA 2x2 two groups two times {Disarmed}

[Agurne Sampedro Calvete]

External Email - Use Caution

Dear Douglas,

Thank you very much for your rapid response. I have another question.

I am following the FS Repeated Measures tutorial and therefore, I ran
this: mris_preproc --target fsaverage --hemi lh --meas thickness --out
lh.thickness.mgh --fsgd rmanova.fsgd.
However, in the  FS Group Analysis tutorial, it says that you must include
"--cache-in thickness.fwhm10.fsaverage" if the data was preprocessed with
recon-all with the -qcache option, which was my case. As I am doing
longitudinal analyses, do I have to skip this  option and just follow the
Repeated Measures tutorial or do I have to include the --cache option?

Thank you very much.
Best regards,

Agurne





Neuropsychology of Severe Medical Conditions Research Team

Dpt of Methods and Experimental Psychology
Faculty of Psychology and Education

University of Deusto

neurolab.deusto.es



Avda. de las Universidades 24,

48007, Bilbao (Spain)

Phone: +34 944139000 (ext. 3015)


El jue., 5 nov. 2020 a las 15:58, Douglas N. Greve ()
escribió:

> In this case, the paired-t is identical to RM 2x2 ANOVA. But to answer
> your question, the first example you give is the right way to handle it.
>
>
> On 11/5/2020 4:32 AM, Agurne Sampedro Calvete wrote:
>
> External Email - Use Caution
> Dear Freesurfer experts,
>
> I am trying to do a longitudinal repeated measures ANOVA 2x2, with two
> groups (experimental vs control) and two time points (pre and post
> treatment).
>
> Although I have run a paired t-test, now I would like to do a repeated
> measures anova 2x2, since I am doing the same analyses with other data in
> other softwares.
>
> I am trying to follow the tutorial *MailScanner has detected a possible
> fraud attempt from "secure-web.cisco.com" claiming to be*
> https://secure-web.cisco.com/11u6-IyY2-P7w7M4zmGiLCyaG6g2vViEpx4ag8H474t_vAcKQofRlrgzoecVmHS-zCmukHiQ4nYncdNoWgv3qseGeeJ5V9DyPr8qfYwFtrgu7s9OZWm-ZxMwuQ5ZFRq0n-rmVcPAZ0Hk-kzFtomVz5K6pII7oYat7YlIYPSB5e0Ig4auXkKYY9vdGnvAE3T5jbnApRosUxoICfXaCSW4-Fw/https%3A%2F%2Fsurfer.nmr.mgh.harvard.edu%2Ffswiki%2FRepeatedMeasuresAnova
> ,
> but I have several doubts when designing the FSGD file as well as the
> contrasts.
>
> How should it be the FSGD file?
>
> I have seen in a previous post the following design for the same number fo
> groups and time points:
>
> GroupDescriptorFile 1
>
> Class Subject1
>
> Class Subject2
>
> Variables Tp1vsTp2.GrExp
>Tp1vsTp2.GrControl
>
> Input   Subject1_time1  Subject1 1
> 0
>
> Input   Subject1_time2  Subject1-10
>
> Input   Subject2_time1  Subject2 0
> 1
>
> Input   Subject2_time2  Subject02   0
> -1
>
>
>
> Design matrix:
>
> 1 0 1 0
>
> 1 0 -1 0
>
> 0 1 0 1
>
> 0 1 0 -1
>
> Contrasts (.mtx files):
> Tp1vsTp2.GrExp0 0 1 0
> Tp1vsTp2.GrControl  0 0 0 -1
> Interaction  0 0 1 -1
>
> Would this design be the correct one to use in this case?
>
> Alternatively, could it be this the correct one?
>
> GroupDescriptorFile 1
>
> Class Subject1
>
> Class Subject2
>
> Variables Tp1vsTp2
>  Groupxtime
>
> Input   Subject1_time1  Subject1 11
>
> Input   Subject1_time2  Subject1-1
> *-1*
>
> Input   Subject2_time1  Subject2 1
> *-1*
>
> Input   Subject2_time2  Subject02   -1  1
>
> Design matrix:
>
> 1 0  1 1
>
> 1 0 -1 -1
>
> 0 1  1 -1
>
> 0 1  -1 1
>
>
>
> Contrasts (.mtx files):
>
> Tp1vsTp20 0 1 0
>
> Groupxtime Interaction   0 0 0 1
>
> Thank you very much in advance.
>
> Kind regards,
>
> Agurne
>
>
>
> Neuropsychology of Severe Medical Conditions Research Team
>
> Dpt of Methods and Experimental Psychology
> Faculty of Psychology and Education
>
> University of Deusto
>
> neurolab.deusto.es
> 
>
>
>
> Avda. de las Universidades 24,
>
> 48007, Bilbao (Spain)
>
> Phone: +34 944139000 (ext. 3015)
>
> ___
> Freesurfer mailing 
> listfreesur...@nmr.mgh.harvard.eduhttps://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
>
>
> ___
> Freesurfer mailing list
> Freesurfer@nmr.mgh.harvard.edu
> 

Re: [Freesurfer] repeated measures ANOVA 2x2 two groups two times {Disarmed}

[Douglas N. Greve]

In this case, the paired-t is identical to RM 2x2 ANOVA. But to answer 
your question, the first example you give is the right way to handle it.



On 11/5/2020 4:32 AM, Agurne Sampedro Calvete wrote:


External Email - Use Caution

Dear Freesurfer experts,

I am trying to do a longitudinal repeated measures ANOVA 2x2, with two 
groups (experimental vs control) and two time points (pre and post 
treatment).


Although I have run a paired t-test, now I would like to do a repeated 
measures anova 2x2, since I am doing the same analyses with other data 
in other softwares.


I am trying to follow the tutorial *MailScanner has detected a 
possible fraud attempt from "secure-web.cisco.com" claiming to be* 
https://surfer.nmr.mgh.harvard.edu/fswiki/RepeatedMeasuresAnova 
, 
but I have several doubts when designing the FSGD file as well as the 
contrasts.


How should it be the FSGD file?

I have seen in a previous post the following design for the same 
number fo groups and time points:


GroupDescriptorFile 1

Class Subject1

Class Subject2

Variables Tp1vsTp2.GrExp 
   Tp1vsTp2.GrControl


Input   Subject1_time1  Subject1 
1 0


Input   Subject1_time2  Subject1    
-1    0


Input   Subject2_time1  Subject2 0 1

Input   Subject2_time2  Subject02  
 0 -1


Design matrix:

1 0 1 0

1 0 -1 0

0 1 0 1

0 1 0 -1


Contrasts (.mtx files):
Tp1vsTp2.GrExp        0 0 1 0
Tp1vsTp2.GrControl  0 0 0 -1
Interaction 0 0 1 -1

Would this design be the correct one to use in this case?

Alternatively, could it be this the correct one?

GroupDescriptorFile 1

Class Subject1

Class Subject2

Variables Tp1vsTp2 
 Groupxtime


Input  Subject1_time1  Subject1 1    1

Input  Subject1_time2  Subject1 -1 *-1*

Input   Subject2_time1  Subject2   1 *-1*

Input  Subject2_time2  Subject02  -1  1

Design matrix:

1 0  1 1

1 0 -1 -1

0 1  1 -1

0 1  -1 1

Contrasts (.mtx files):

Tp1vsTp2 0 0 1 0

Groupxtime Interaction   0 0 0 1


Thank you very much in advance.

Kind regards,

Agurne

Neuropsychology of Severe Medical Conditions Research Team

Dpt of Methods and Experimental Psychology
Faculty of Psychology and Education

University of Deusto

neurolab.deusto.es 



Avda. de las Universidades 24,

48007, Bilbao (Spain)

Phone: +34 944139000 (ext. 3015)


___
Freesurfer mailing list
Freesurfer@nmr.mgh.harvard.edu
https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer


___
Freesurfer mailing list
Freesurfer@nmr.mgh.harvard.edu
https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer

[Freesurfer] repeated measures ANOVA 2x2 two groups two times {Disarmed}

[Agurne Sampedro Calvete]

External Email - Use Caution

Dear Freesurfer experts,

I am trying to do a longitudinal repeated measures ANOVA 2x2, with two
groups (experimental vs control) and two time points (pre and post
treatment).

Although I have run a paired t-test, now I would like to do a repeated
measures anova 2x2, since I am doing the same analyses with other data in
other softwares.

I am trying to follow the tutorial
https://secure-web.cisco.com/1xcEvqAuPN3MsqK1JjC6qoZbooZS3kpB4WEU1Lx8hK28grFkOWFgj7ja5TXppcZRh8Vr2s3TJyQKT-sr-c0qt4XJSU7f74cElgVAI4K1nOBI5i5L0YQBzYutTVn-MwCGM_OotAt9VM1mzmm1MUuXbYtZC4-QfIujBH30LAwD5RtwZ6vGRv4JC7ocISixV0s52BsCadrLE2J-3CdqLLZnJRA/https%3A%2F%2Fsurfer.nmr.mgh.harvard.edu%2Ffswiki%2FRepeatedMeasuresAnova,
 but I have
several doubts when designing the FSGD file as well as the contrasts.

How should it be the FSGD file?

I have seen in a previous post the following design for the same number fo
groups and time points:

GroupDescriptorFile 1

Class Subject1

Class Subject2

Variables Tp1vsTp2.GrExp
   Tp1vsTp2.GrControl

Input   Subject1_time1  Subject1 1 0

Input   Subject1_time2  Subject1-10

Input   Subject2_time1  Subject2 0
1

Input   Subject2_time2  Subject02   0 -1



Design matrix:

1 0 1 0

1 0 -1 0

0 1 0 1

0 1 0 -1

Contrasts (.mtx files):
Tp1vsTp2.GrExp0 0 1 0
Tp1vsTp2.GrControl  0 0 0 -1
Interaction  0 0 1 -1

Would this design be the correct one to use in this case?

Alternatively, could it be this the correct one?

GroupDescriptorFile 1

Class Subject1

Class Subject2

Variables Tp1vsTp2
 Groupxtime

Input   Subject1_time1  Subject1 11

Input   Subject1_time2  Subject1-1
*-1*

Input   Subject2_time1  Subject2 1
*-1*

Input   Subject2_time2  Subject02   -1  1

Design matrix:

1 0  1 1

1 0 -1 -1

0 1  1 -1

0 1  -1 1



Contrasts (.mtx files):

Tp1vsTp20 0 1 0

Groupxtime Interaction   0 0 0 1

Thank you very much in advance.

Kind regards,

Agurne



Neuropsychology of Severe Medical Conditions Research Team

Dpt of Methods and Experimental Psychology
Faculty of Psychology and Education

University of Deusto

neurolab.deusto.es



Avda. de las Universidades 24,

48007, Bilbao (Spain)

Phone: +34 944139000 (ext. 3015)
___
Freesurfer mailing list
Freesurfer@nmr.mgh.harvard.edu
https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer

Re: [Freesurfer] repeated measures ANOVA with two groups

[Douglas N Greve]

Hi Shannon, sorry for the delay. There are several ways you can do this. 
If you already have a slope for each subject, then just use that as 
input and use the simple two group FSGD file. If you really want to do 
the RM ANOVA, then there are two ways. The simpler way is to create a 
stack of all subjects for each visit (eg, v1, v2, and v3) using 
mris_preproc. Next compute paired differences d12 = v1-v2 and d13 = 
v1-v3 using fscalc (note that you do not need v2-v3). Next, create a new 
stack from d12 and d13 with mri_concat (eg, mri_concat d12.mgh d13.mgh 
--o d12+d13.mgh). Then create an FSGD file with a class for each group 
and two variables (d12 and d13)). The structure will look something like 
this:

Input subject1 G1 1 0
Input subject2 G1 1 0
... rest of subjects for group 1 ...
Input subject1 G1 0 1
Input subject2 G1 0 1
... rest of subjects for group 1 ...
Input subject1 G2 1 0
Input subject2 G2 1 0
... rest of subjects for group 2 ...
Input subject1 G2 0 1
Input subject2 G2 0 1
... rest of subjects for group 2 ...

Run the analysis with DOSS. You can use the following contrast matrix to 
test for an interaction:
[1 -1 -1 1]

Let me know if that makes sense or not
doug



On 08/23/2012 09:39 PM, Shannon Kogachi wrote:
 Hi,
   Is there a way to set up a repeated measures ANOVA and ANCOVA with 
 two different groups (ex: study group v. controls) and each subject 
 has more than two timepoints? Would it be possible to assign each 
 Input in the fsgd file to two classes (ex: timepoint scan and group) 
 or is there another way to set up the fsgd file and contrast matrices? 
 I've already used qdec in freesurfer 5.1 to calculate rates of change 
 and symmetrized percent changes between the timepoints (using 
 long_mris_slopes) but now would like to perform a repeated measures 
 test. Thank you!

 -Shannon K.


 ___
 Freesurfer mailing list
 Freesurfer@nmr.mgh.harvard.edu
 https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer

-- 
Douglas N. Greve, Ph.D.
MGH-NMR Center
gr...@nmr.mgh.harvard.edu
Phone Number: 617-724-2358
Fax: 617-726-7422

Bugs: surfer.nmr.mgh.harvard.edu/fswiki/BugReporting
FileDrop: www.nmr.mgh.harvard.edu/facility/filedrop/index.html

___
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] repeated measures ANOVA with two groups

[Shannon Kogachi]

Hi,
  Is there a way to set up a repeated measures ANOVA and ANCOVA with two
different groups (ex: study group v. controls) and each subject has more
than two timepoints? Would it be possible to assign each Input in the fsgd
file to two classes (ex: timepoint scan and group) or is there another way
to set up the fsgd file and contrast matrices? I've already used qdec in
freesurfer 5.1 to calculate rates of change and symmetrized percent changes
between the timepoints (using long_mris_slopes) but now would like to
perform a repeated measures test. Thank you!

-Shannon K.
___
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.