Re: [Freesurfer] beta weights from FS-Fast analysis

[Douglas N Greve]

Oh, right, the nuisance regressors are given a separate regressor for 
each run. I had forgotten!

On 07/17/2013 04:55 PM, Joseph Dien wrote:
>
> On Jul 17, 2013, at 4:50 PM, Douglas N Greve 
> mailto:gr...@nmr.mgh.harvard.edu>> wrote:
>
>>
>> On 07/17/2013 04:48 PM, Joseph Dien wrote:
>>> The single nuisance regressor models the difference between the merged
>>> runs.  So if the first run had a mean of 90 and the second run had a
>>> mean of 110 then the merged run mean would be 100.  The nuisance
>>> regressor has 1s for the volumes of the first run and -1s for the
>>> volumes of the second run so it ends up with a beta value of 10, thus
>>> accounting for the difference between these two sets of volumes.  I
>>> think it does make sense.
>> Do you do this in a single regressor? So you would have a +1 -1 pattern
>> repeated 4 times? I think to make it work, you would need 4 regressors.
>> In any event, FSFAST will do the right thing, so maybe it is not 
>> important.
>>>
>
> Looking at the X.X file, it did create four nuisance regressors.  I'm 
> really impressed with how well FSFAST handles all this!  :)
>
>
>>> In any case, while it was necessary to do so for my original SPM
>>> analyses since it uses a separate covariates for each run, after
>>> working through the FSFAST procedures with your help I see that is not
>>> the case for FSFAST (a single regressor models a given condition in
>>> all the runs).  I'll try doing as you suggest to see what difference
>>> it makes.
>>>
>>> It is indeed cognitive areas and the manipulations are subtle social
>>> cognition manipulations so perhaps not surprising after all.
>>>
>>> I'll send you the paradigm file separately.
>>>
>>> Thanks for taking the time to look into this!
>>>
>>> Joe
>>>
>>>
>>> On Jul 17, 2013, at 4:38 PM, Douglas N Greve
>>> mailto:gr...@nmr.mgh.harvard.edu> 
>>> > wrote:
>>>

 It is not necessary or beneficial to combine the runs in this way.
 FSFAST will do all this for you and keep account of all the runs and
 transitions. FSFAST will put in regressors to fit each of the run 
 means.
 The single regressor you have is not the right way to hand this (at
 least I don't understand how it works). It could be that the low %
 signal change is related to the colinearity between the task waveform
 and the mean regressors. Can you set things up in the way that FSFAST
 expects them and don't use any nuisance regressors?

 Also, in what area are you looking at the percent change? .02% sounds
 very small, but may be if it is in some cognitive area, maybe it is ok.
 If it is in visual cortex, then it looks way too low.

 Also, can you send the paradigm file?

 doug




 On 07/17/2013 04:27 PM, Joseph Dien wrote:
> It's a little complicated.  Basically there were eight runs,
> comprising four conditions (me, we, you1, you2) each with two
> adjoining runs.  For the analysis, I merged each of the pairs into a
> single run and added a nuisance regressor to account for the
> difference in run means.  There were a total of four different kinds
> of boxcars (AR, CS, EM, MP).  So 4x4=16 conditions.  There was also a
> covariate of non-interest to mark the switch point for each boxcar,
> one for each run, so 20 total.
>
> The 7 nuisance regressors are six movement covariates plus one to
> account for merging eight runs into four (it consists of 1 for the
> first half and -1 for the second, so the difference in the run means).
> I'm using the movement covariates from a prior SPM run since
> ARTdetect (for detecting bad volumes) isn't set up for AFNI style
> data.  From all published accounts the different movement detection
> routines yield similar enough results that it shouldn't be a problem
> (consistent with what I found when I compared them for this dataset).
>
> You're thinking that collinearity could have reduced the effect sizes?
> When I correlate the X.X regressor matrix, the 20 predictors don't
> correlate by more than about .2 at worst.  I do see greater
> correlations with some of the nuisance regressors (as high as the .4
> range).  Are my betas unusually small for FSFAST analyses?  They did
> come up clusterwise significant at least. Or should I not worry?  I'm
> not sure what to expect from FSFAST analyses.
>
> Thanks!
>
> Joe
>
>
> On Jul 17, 2013, at 3:58 PM, Douglas N Greve
> mailto:gr...@nmr.mgh.harvard.edu> 
> 
> > wrote:
>
>> why do you have 20 conditions? And what are the 7 nuisance 
>> regressors?
>>
>> On 07/17/2013 03:54 PM, Joseph Dien wrote:
>>> It's a boxcar design so 20.265.
>>>
>>> mkanalysis-sess -fsd bold -analysis CPA.sm05.lh -surface 
>>> fsaverage lh
>>> -fwhm

Re: [Freesurfer] beta weights from FS-Fast analysis

[Joseph Dien]

It's a little complicated.  Basically there were eight runs, comprising four 
conditions (me, we, you1, you2) each with two adjoining runs.  For the 
analysis, I merged each of the pairs into a single run and added a nuisance 
regressor to account for the difference in run means.  There were a total of 
four different kinds of boxcars (AR, CS, EM, MP).  So 4x4=16 conditions.  There 
was also a covariate of non-interest to mark the switch point for each boxcar, 
one for each run, so 20 total.

The 7 nuisance regressors are six movement covariates plus one to account for 
merging eight runs into four (it consists of 1 for the first half and -1 for 
the second, so the difference in the run means).  I'm using the movement 
covariates from a prior SPM run since ARTdetect (for detecting bad volumes) 
isn't set up for AFNI style data.  From all published accounts the different 
movement detection routines yield similar enough results that it shouldn't be a 
problem (consistent with what I found when I compared them for this dataset).

You're thinking that collinearity could have reduced the effect sizes?  When I 
correlate the X.X regressor matrix, the 20 predictors don't correlate by more 
than about .2 at worst.  I do see greater correlations with some of the 
nuisance regressors (as high as the .4 range).  Are my betas unusually small 
for FSFAST analyses?  They did come up clusterwise significant at least. Or 
should I not worry?  I'm not sure what to expect from FSFAST analyses.

Thanks!

Joe


On Jul 17, 2013, at 3:58 PM, Douglas N Greve  wrote:

> why do you have 20 conditions? And what are the 7 nuisance regressors?
> 
> On 07/17/2013 03:54 PM, Joseph Dien wrote:
>> It's a boxcar design so 20.265.
>> 
>> mkanalysis-sess -fsd bold -analysis CPA.sm05.lh -surface fsaverage lh 
>> -fwhm 5 -event-related-paradigm CPA1.par -nconditions 20 -spmhrf 0 -TR 
>> 2 -refeventdur 20.265 -polyfit 2 -per-run -force -nuisreg nuisreg2.dat 
>> 7 -tpexclude tpexclude.dat
>> 
>> On Jul 17, 2013, at 3:50 PM, Douglas N Greve 
>> mailto:gr...@nmr.mgh.harvard.edu>> wrote:
>> 
>>> when you ran mkanalysis-sess, what did you set --refeventdur to?
>>> On 07/17/2013 02:50 PM, Joseph Dien wrote:
 then I get on the order of .02% difference between the contrasted 
 conditions.
 The run mean values are in my expected ballpark of about 100 or so.
 The condition betas are just very very small.
 Or perhaps this is typical of FSFAST analyses?
 
 On Jul 17, 2013, at 2:00 PM, Douglas N Greve 
 mailto:gr...@nmr.mgh.harvard.edu> 
 > wrote:
 
> 
> The beta's have already been scaled. What do you get if you just
> beta/runmean ?
> 
> 
> 
> On 07/17/2013 01:45 PM, Joseph Dien wrote:
>> I implemented the ROI percent signal change formula following the
>> MarsBaR FAQ (http://marsbar.sourceforge.net/faq.html) but the values
>> I'm getting seem too small (on the order of .0002%).  Basically the
>> formula is the (beta * peak absolute value of the canonical HRF
>> regressor * 100)/(run mean).  No derivatives in this case as it is a
>> boxcar design.
>> 
>> I took the mean across all the runs since FSFAST uses the same
>> regressor across the entire experiment (unlike SPM).
>> I used the X.runflac(1).flac.ev(m).Xirf values for the canonical HRF
>> as you suggested (where m equals the condition+1).
>> 
>> Is it possible that I'm missing something in the scaling here?
>> Especially with a boxcar design, the signal change should be much
>> larger than this for a significant cluster, I think.  For example, the
>> peak HRF value for one of the conditions is 0.0092.  If the betas are
>> already scaled according to the peak value, then it would come out as
>> .02%, which is more reasonable, although still too small.
>> 
>> Thanks for your help with this!
>> 
>> Joe
>> 
>> 
>> 
>> On May 31, 2013, at 5:02 PM, Douglas N Greve
>> mailto:gr...@nmr.mgh.harvard.edu> 
>>  
>> > wrote:
>> 
>>> 
>>> Oh, right, it is probably not there for subcortical. I don't know 
>>> what I
>>> would have to do to write it out. It won't be something that happens
>>> before I get back from HBM. Can you remind me after HBM?
>>> doug
>>> 
>>> On 05/31/2013 04:44 PM, Joseph Dien wrote:
 It looks like the corrected vertex p-values
 (ex: cache.th13.abs.sig.voxel.nii.gz) are only available for the
 surface-based lh and rh spaces.  For the subcortical volume-based
 analysis I don't see the corresponding corrected voxel p-values 
 being
 available?
 
 On May 31, 2013, at 2:46 PM, Joseph Dien >>>  
 
 > 

Re: [Freesurfer] beta weights from FS-Fast analysis

[Joseph Dien]

It's a boxcar design so 20.265.

  mkanalysis-sess -fsd bold -analysis CPA.sm05.lh -surface fsaverage lh 
-fwhm 5 -event-related  -paradigm CPA1.par -nconditions 20 -spmhrf 0 -TR 2 
-refeventdur 20.265 -polyfit 2 -per-run -force -nuisreg nuisreg2.dat 7 
-tpexclude tpexclude.dat

On Jul 17, 2013, at 3:50 PM, Douglas N Greve  wrote:

> when you ran mkanalysis-sess, what did you set --refeventdur to?
> On 07/17/2013 02:50 PM, Joseph Dien wrote:
>> then I get on the order of .02% difference between the contrasted conditions.
>> The run mean values are in my expected ballpark of about 100 or so.
>> The condition betas are just very very small.
>> Or perhaps this is typical of FSFAST analyses?
>> 
>> On Jul 17, 2013, at 2:00 PM, Douglas N Greve > > wrote:
>> 
>>> 
>>> The beta's have already been scaled. What do you get if you just
>>> beta/runmean ?
>>> 
>>> 
>>> 
>>> On 07/17/2013 01:45 PM, Joseph Dien wrote:
 I implemented the ROI percent signal change formula following the
 MarsBaR FAQ (http://marsbar.sourceforge.net/faq.html) but the values
 I'm getting seem too small (on the order of .0002%).  Basically the
 formula is the (beta * peak absolute value of the canonical HRF
 regressor * 100)/(run mean).  No derivatives in this case as it is a
 boxcar design.
 
 I took the mean across all the runs since FSFAST uses the same
 regressor across the entire experiment (unlike SPM).
 I used the X.runflac(1).flac.ev(m).Xirf values for the canonical HRF
 as you suggested (where m equals the condition+1).
 
 Is it possible that I'm missing something in the scaling here?
 Especially with a boxcar design, the signal change should be much
 larger than this for a significant cluster, I think.  For example, the
 peak HRF value for one of the conditions is 0.0092.  If the betas are
 already scaled according to the peak value, then it would come out as
 .02%, which is more reasonable, although still too small.
 
 Thanks for your help with this!
 
 Joe
 
 
 
 On May 31, 2013, at 5:02 PM, Douglas N Greve
 mailto:gr...@nmr.mgh.harvard.edu> 
 > wrote:
 
> 
> Oh, right, it is probably not there for subcortical. I don't know what I
> would have to do to write it out. It won't be something that happens
> before I get back from HBM. Can you remind me after HBM?
> doug
> 
> On 05/31/2013 04:44 PM, Joseph Dien wrote:
>> It looks like the corrected vertex p-values
>> (ex: cache.th13.abs.sig.voxel.nii.gz) are only available for the
>> surface-based lh and rh spaces.  For the subcortical volume-based
>> analysis I don't see the corresponding corrected voxel p-values being
>> available?
>> 
>> On May 31, 2013, at 2:46 PM, Joseph Dien > 
>> 
>> > wrote:
>> 
>>> 
>>> On May 31, 2013, at 12:11 PM, Douglas N Greve
>>> mailto:gr...@nmr.mgh.harvard.edu> 
>>> 
>>> > wrote:
>>> 
 
 On 05/31/2013 01:49 AM, Joseph Dien wrote:
> I was able to make more progress so I'm mostly good at this point but
> I have a remaining question:
> 
> I assume the contents of sig.nii.gz (which I assume are the vertex
> p-values) are not FWE corrected.  Is it possible to get FWE-corrected
> vertex p-values?  Or are only clusterwise corrections available?
 There should be something like cache.th13.abs.sig.voxel.mgh which is
 corrected on a voxelwise basis (the th13 is just part of the name
 but it
 should be the same regardless of the threshold you choose)
 doug
>>> 
>>> Excellent!  Thanks!  :)
>>> 
> 
> Thanks again for your patience!
> 
> Joe
> 
> On May 30, 2013, at 4:37 PM, Joseph Dien  
> 
> 
> > wrote:
> 
>> Just to make sure I'm doing this right, I'm going to summarize what
>> I've taken away from your answers and to ask some new questions. In
>> order to present the results, I need two things:
>> 
>> 1) A set of histograms (with error bars) for each cluster figure to
>> show the % signal change for each of the four contrasts of interest.
>> The cache.th20.pos.y.ocn.dat file only gives it for the condition
>> where the cluster was significant so I can't use that.
>> So I could use mri_label2vol to convert cache.th20.neg.sig.ocn.annot
>> from the group level analysis to generate a mask for each cluster of
>> interest.
>> Then I could extract the value of the vo

Re: [Freesurfer] beta weights from FS-Fast analysis

[Joseph Dien]

then I get on the order of .02% difference between the contrasted conditions.
The run mean values are in my expected ballpark of about 100 or so.
The condition betas are just very very small.
Or perhaps this is typical of FSFAST analyses?

On Jul 17, 2013, at 2:00 PM, Douglas N Greve  wrote:

> 
> The beta's have already been scaled. What do you get if you just 
> beta/runmean ?
> 
> 
> 
> On 07/17/2013 01:45 PM, Joseph Dien wrote:
>> I implemented the ROI percent signal change formula following the 
>> MarsBaR FAQ (http://marsbar.sourceforge.net/faq.html) but the values 
>> I'm getting seem too small (on the order of .0002%).  Basically the 
>> formula is the (beta * peak absolute value of the canonical HRF 
>> regressor * 100)/(run mean).  No derivatives in this case as it is a 
>> boxcar design.
>> 
>> I took the mean across all the runs since FSFAST uses the same 
>> regressor across the entire experiment (unlike SPM).
>> I used the X.runflac(1).flac.ev(m).Xirf values for the canonical HRF 
>> as you suggested (where m equals the condition+1).
>> 
>> Is it possible that I'm missing something in the scaling here? 
>> Especially with a boxcar design, the signal change should be much 
>> larger than this for a significant cluster, I think.  For example, the 
>> peak HRF value for one of the conditions is 0.0092.  If the betas are 
>> already scaled according to the peak value, then it would come out as 
>> .02%, which is more reasonable, although still too small.
>> 
>> Thanks for your help with this!
>> 
>> Joe
>> 
>> 
>> 
>> On May 31, 2013, at 5:02 PM, Douglas N Greve 
>> mailto:gr...@nmr.mgh.harvard.edu>> wrote:
>> 
>>> 
>>> Oh, right, it is probably not there for subcortical. I don't know what I
>>> would have to do to write it out. It won't be something that happens
>>> before I get back from HBM. Can you remind me after HBM?
>>> doug
>>> 
>>> On 05/31/2013 04:44 PM, Joseph Dien wrote:
 It looks like the corrected vertex p-values
 (ex: cache.th13.abs.sig.voxel.nii.gz) are only available for the
 surface-based lh and rh spaces.  For the subcortical volume-based
 analysis I don't see the corresponding corrected voxel p-values being
 available?
 
 On May 31, 2013, at 2:46 PM, Joseph Dien >>> 
 > wrote:
 
> 
> On May 31, 2013, at 12:11 PM, Douglas N Greve
> mailto:gr...@nmr.mgh.harvard.edu> 
> > wrote:
> 
>> 
>> On 05/31/2013 01:49 AM, Joseph Dien wrote:
>>> I was able to make more progress so I'm mostly good at this point but
>>> I have a remaining question:
>>> 
>>> I assume the contents of sig.nii.gz (which I assume are the vertex
>>> p-values) are not FWE corrected.  Is it possible to get FWE-corrected
>>> vertex p-values?  Or are only clusterwise corrections available?
>> There should be something like cache.th13.abs.sig.voxel.mgh which is
>> corrected on a voxelwise basis (the th13 is just part of the name
>> but it
>> should be the same regardless of the threshold you choose)
>> doug
> 
> Excellent!  Thanks!  :)
> 
>>> 
>>> Thanks again for your patience!
>>> 
>>> Joe
>>> 
>>> On May 30, 2013, at 4:37 PM, Joseph Dien >> 
>>> 
>>> > wrote:
>>> 
 Just to make sure I'm doing this right, I'm going to summarize what
 I've taken away from your answers and to ask some new questions. In
 order to present the results, I need two things:
 
 1) A set of histograms (with error bars) for each cluster figure to
 show the % signal change for each of the four contrasts of interest.
 The cache.th20.pos.y.ocn.dat file only gives it for the condition
 where the cluster was significant so I can't use that.
 So I could use mri_label2vol to convert cache.th20.neg.sig.ocn.annot
 from the group level analysis to generate a mask for each cluster of
 interest.
 Then I could extract the value of the voxels from each
 subject's cespct file for each contrast, average them across the
 cluster ROI, then average them across each subject, to generate the
 histogram?
 This would suffice to give me the %age signal change?
 I would be doing these computations in Matlab using MRIread.
 
 2) A results table with the headings:
 
 Cluster p (FWE corrected)
 Cluster size
 Peak Voxel p (FWE corrected)
 Peak Voxel T
 Peak Voxel Coords
 BA
 Anatomical Landmark
 
 I can get the first two from
 the cache.th20.pos/neg.sig.cluster.summary files from the group 
 level
 analysis.
 I can get the peak voxel coordinates from the summary files as well.
 I can use this to get

Re: [Freesurfer] beta weights from FS-Fast analysis

[Joseph Dien]

I implemented the ROI percent signal change formula following the MarsBaR FAQ 
(http://marsbar.sourceforge.net/faq.html) but the values I'm getting seem too 
small (on the order of .0002%).  Basically the formula is the (beta * peak 
absolute value of the canonical HRF regressor * 100)/(run mean).  No 
derivatives in this case as it is a boxcar design.

I took the mean across all the runs since FSFAST uses the same regressor across 
the entire experiment (unlike SPM).
I used the X.runflac(1).flac.ev(m).Xirf values for the canonical HRF as you 
suggested (where m equals the condition+1). 

Is it possible that I'm missing something in the scaling here?  Especially with 
a boxcar design, the signal change should be much larger than this for a 
significant cluster, I think.  For example, the peak HRF value for one of the 
conditions is 0.0092.  If the betas are already scaled according to the peak 
value, then it would come out as .02%, which is more reasonable, although still 
too small.

Thanks for your help with this!

Joe



On May 31, 2013, at 5:02 PM, Douglas N Greve  wrote:

> 
> Oh, right, it is probably not there for subcortical. I don't know what I 
> would have to do to write it out. It won't be something that happens 
> before I get back from HBM. Can you remind me after HBM?
> doug
> 
> On 05/31/2013 04:44 PM, Joseph Dien wrote:
>> It looks like the corrected vertex p-values 
>> (ex: cache.th13.abs.sig.voxel.nii.gz) are only available for the 
>> surface-based lh and rh spaces.  For the subcortical volume-based 
>> analysis I don't see the corresponding corrected voxel p-values being 
>> available?
>> 
>> On May 31, 2013, at 2:46 PM, Joseph Dien > > wrote:
>> 
>>> 
>>> On May 31, 2013, at 12:11 PM, Douglas N Greve 
>>> mailto:gr...@nmr.mgh.harvard.edu>> wrote:
>>> 
 
 On 05/31/2013 01:49 AM, Joseph Dien wrote:
> I was able to make more progress so I'm mostly good at this point but
> I have a remaining question:
> 
> I assume the contents of sig.nii.gz (which I assume are the vertex
> p-values) are not FWE corrected.  Is it possible to get FWE-corrected
> vertex p-values?  Or are only clusterwise corrections available?
 There should be something like cache.th13.abs.sig.voxel.mgh which is
 corrected on a voxelwise basis (the th13 is just part of the name 
 but it
 should be the same regardless of the threshold you choose)
 doug
>>> 
>>> Excellent!  Thanks!  :)
>>> 
> 
> Thanks again for your patience!
> 
> Joe
> 
> On May 30, 2013, at 4:37 PM, Joseph Dien  
> > wrote:
> 
>> Just to make sure I'm doing this right, I'm going to summarize what
>> I've taken away from your answers and to ask some new questions. In
>> order to present the results, I need two things:
>> 
>> 1) A set of histograms (with error bars) for each cluster figure to
>> show the % signal change for each of the four contrasts of interest.
>> The cache.th20.pos.y.ocn.dat file only gives it for the condition
>> where the cluster was significant so I can't use that.
>> So I could use mri_label2vol to convert cache.th20.neg.sig.ocn.annot
>> from the group level analysis to generate a mask for each cluster of
>> interest.
>> Then I could extract the value of the voxels from each
>> subject's cespct file for each contrast, average them across the
>> cluster ROI, then average them across each subject, to generate the
>> histogram?
>> This would suffice to give me the %age signal change?
>> I would be doing these computations in Matlab using MRIread.
>> 
>> 2) A results table with the headings:
>> 
>> Cluster p (FWE corrected)
>> Cluster size
>> Peak Voxel p (FWE corrected)
>> Peak Voxel T
>> Peak Voxel Coords
>> BA
>> Anatomical Landmark
>> 
>> I can get the first two from
>> the cache.th20.pos/neg.sig.cluster.summary files from the group level
>> analysis.
>> I can get the peak voxel coordinates from the summary files as well.
>> I can use this to get the peak voxel p from the group
>> level sig.nii.gz file.  Is this FWE corrected?  If not, how can I get
>> this information?
>> I can use these coordinates to get the peak voxel T by getting the
>> value from the group level F.nii.gz file and taking its square root.
>> How can I get the sign of the T statistic?
>> I can use the Lancaster transform to convert the MNI305 peak voxel
>> coordinates into the Atlas coordinates to look up the putative BA and
>> landmarks (unless there is a better way with Freesurfer?  I'm seeing
>> some references to some BA labels in the forum but it doesn't look
>> like this is a complete set yet?).
>> 
>> Sorry for all these questions!  I got some nice results from FSFAST
>> and would like to get them written up.
>> 
>>

Re: [Freesurfer] beta weights from FS-Fast analysis

[Douglas N Greve]

Oh, right, it is probably not there for subcortical. I don't know what I 
would have to do to write it out. It won't be something that happens 
before I get back from HBM. Can you remind me after HBM?
doug

On 05/31/2013 04:44 PM, Joseph Dien wrote:
> It looks like the corrected vertex p-values 
> (ex: cache.th13.abs.sig.voxel.nii.gz) are only available for the 
> surface-based lh and rh spaces.  For the subcortical volume-based 
> analysis I don't see the corresponding corrected voxel p-values being 
> available?
>
> On May 31, 2013, at 2:46 PM, Joseph Dien  > wrote:
>
>>
>> On May 31, 2013, at 12:11 PM, Douglas N Greve 
>> mailto:gr...@nmr.mgh.harvard.edu>> wrote:
>>
>>>
>>> On 05/31/2013 01:49 AM, Joseph Dien wrote:
 I was able to make more progress so I'm mostly good at this point but
 I have a remaining question:

 I assume the contents of sig.nii.gz (which I assume are the vertex
 p-values) are not FWE corrected.  Is it possible to get FWE-corrected
 vertex p-values?  Or are only clusterwise corrections available?
>>> There should be something like cache.th13.abs.sig.voxel.mgh which is
>>> corrected on a voxelwise basis (the th13 is just part of the name 
>>> but it
>>> should be the same regardless of the threshold you choose)
>>> doug
>>
>> Excellent!  Thanks!  :)
>>

 Thanks again for your patience!

 Joe

 On May 30, 2013, at 4:37 PM, Joseph Dien >>> 
 > wrote:

> Just to make sure I'm doing this right, I'm going to summarize what
> I've taken away from your answers and to ask some new questions. In
> order to present the results, I need two things:
>
> 1) A set of histograms (with error bars) for each cluster figure to
> show the % signal change for each of the four contrasts of interest.
> The cache.th20.pos.y.ocn.dat file only gives it for the condition
> where the cluster was significant so I can't use that.
> So I could use mri_label2vol to convert cache.th20.neg.sig.ocn.annot
> from the group level analysis to generate a mask for each cluster of
> interest.
> Then I could extract the value of the voxels from each
> subject's cespct file for each contrast, average them across the
> cluster ROI, then average them across each subject, to generate the
> histogram?
> This would suffice to give me the %age signal change?
> I would be doing these computations in Matlab using MRIread.
>
> 2) A results table with the headings:
>
> Cluster p (FWE corrected)
> Cluster size
> Peak Voxel p (FWE corrected)
> Peak Voxel T
> Peak Voxel Coords
> BA
> Anatomical Landmark
>
> I can get the first two from
> the cache.th20.pos/neg.sig.cluster.summary files from the group level
> analysis.
> I can get the peak voxel coordinates from the summary files as well.
> I can use this to get the peak voxel p from the group
> level sig.nii.gz file.  Is this FWE corrected?  If not, how can I get
> this information?
> I can use these coordinates to get the peak voxel T by getting the
> value from the group level F.nii.gz file and taking its square root.
> How can I get the sign of the T statistic?
> I can use the Lancaster transform to convert the MNI305 peak voxel
> coordinates into the Atlas coordinates to look up the putative BA and
> landmarks (unless there is a better way with Freesurfer?  I'm seeing
> some references to some BA labels in the forum but it doesn't look
> like this is a complete set yet?).
>
> Sorry for all these questions!  I got some nice results from FSFAST
> and would like to get them written up.
>
> Cheers!
>
> Joe
>
>
>
>
> On May 29, 2013, at 10:53 PM, Douglas Greve
> mailto:gr...@nmr.mgh.harvard.edu> 
> > wrote:
>
>>
>> On 5/29/13 10:42 PM, Joseph Dien wrote:
>>>
>>> On May 29, 2013, at 11:40 AM, Douglas N Greve
>>> mailto:gr...@nmr.mgh.harvard.edu> 
>>> > wrote:
>>>
 Hi Joe,

 On 05/29/2013 01:00 AM, Joseph Dien wrote:
> I need to extract the beta weights from a cluster identified with
> FS-Fast in order to compute percentage signal change.
>
> 1) I see a file called beta.nii.gz that appears to have the beta
> weight information.  It has a four dimensional structure and the
> fourth dimension appears to be the beta weights.  Is there an 
> index
> somewhere as to which beta weight is which?  Or if not, how 
> are they
> organized?
 For the first level analysis, the first N beta weights correspond
 to the
 N conditions in the paradigm file. The rest are nuisance variables.
>
>>>
>>> Ah, very good!  In order

Re: [Freesurfer] beta weights from FS-Fast analysis

[Joseph Dien]

It looks like the corrected vertex p-values (ex: 
cache.th13.abs.sig.voxel.nii.gz) are only available for the surface-based lh 
and rh spaces.  For the subcortical volume-based analysis I don't see the 
corresponding corrected voxel p-values being available?

On May 31, 2013, at 2:46 PM, Joseph Dien  wrote:

> 
> On May 31, 2013, at 12:11 PM, Douglas N Greve  
> wrote:
> 
>> 
>> On 05/31/2013 01:49 AM, Joseph Dien wrote:
>>> I was able to make more progress so I'm mostly good at this point but 
>>> I have a remaining question:
>>> 
>>> I assume the contents of sig.nii.gz (which I assume are the vertex 
>>> p-values) are not FWE corrected.  Is it possible to get FWE-corrected 
>>> vertex p-values?  Or are only clusterwise corrections available?
>> There should be something like cache.th13.abs.sig.voxel.mgh which is 
>> corrected on a voxelwise basis (the th13 is just part of the name but it 
>> should be the same regardless of the threshold you choose)
>> doug
> 
> Excellent!  Thanks!  :)
> 
>>> 
>>> Thanks again for your patience!
>>> 
>>> Joe
>>> 
>>> On May 30, 2013, at 4:37 PM, Joseph Dien >> > wrote:
>>> 
 Just to make sure I'm doing this right, I'm going to summarize what 
 I've taken away from your answers and to ask some new questions. In 
 order to present the results, I need two things:
 
 1) A set of histograms (with error bars) for each cluster figure to 
 show the % signal change for each of the four contrasts of interest.
 The cache.th20.pos.y.ocn.dat file only gives it for the condition 
 where the cluster was significant so I can't use that.
 So I could use mri_label2vol to convert cache.th20.neg.sig.ocn.annot 
 from the group level analysis to generate a mask for each cluster of 
 interest.
 Then I could extract the value of the voxels from each 
 subject's cespct file for each contrast, average them across the 
 cluster ROI, then average them across each subject, to generate the 
 histogram?
 This would suffice to give me the %age signal change?
 I would be doing these computations in Matlab using MRIread.
 
 2) A results table with the headings:
 
 Cluster p (FWE corrected)
 Cluster size
 Peak Voxel p (FWE corrected)
 Peak Voxel T
 Peak Voxel Coords
 BA
 Anatomical Landmark
 
 I can get the first two from 
 the cache.th20.pos/neg.sig.cluster.summary files from the group level 
 analysis.
 I can get the peak voxel coordinates from the summary files as well.
 I can use this to get the peak voxel p from the group 
 level sig.nii.gz file.  Is this FWE corrected?  If not, how can I get 
 this information?
 I can use these coordinates to get the peak voxel T by getting the 
 value from the group level F.nii.gz file and taking its square root. 
 How can I get the sign of the T statistic?
 I can use the Lancaster transform to convert the MNI305 peak voxel 
 coordinates into the Atlas coordinates to look up the putative BA and 
 landmarks (unless there is a better way with Freesurfer?  I'm seeing 
 some references to some BA labels in the forum but it doesn't look 
 like this is a complete set yet?).
 
 Sorry for all these questions!  I got some nice results from FSFAST 
 and would like to get them written up.
 
 Cheers!
 
 Joe
 
 
 
 
 On May 29, 2013, at 10:53 PM, Douglas Greve 
 mailto:gr...@nmr.mgh.harvard.edu>> wrote:
 
> 
> On 5/29/13 10:42 PM, Joseph Dien wrote:
>> 
>> On May 29, 2013, at 11:40 AM, Douglas N Greve 
>> mailto:gr...@nmr.mgh.harvard.edu>> wrote:
>> 
>>> Hi Joe,
>>> 
>>> On 05/29/2013 01:00 AM, Joseph Dien wrote:
 I need to extract the beta weights from a cluster identified with
 FS-Fast in order to compute percentage signal change.
 
 1) I see a file called beta.nii.gz that appears to have the beta
 weight information.  It has a four dimensional structure and the
 fourth dimension appears to be the beta weights.  Is there an index
 somewhere as to which beta weight is which?  Or if not, how are they
 organized?
>>> For the first level analysis, the first N beta weights correspond 
>>> to the
>>> N conditions in the paradigm file. The rest are nuisance variables.
 
>> 
>> Ah, very good!  In order to compute the percent signal change 
>> statistic (I'm following the MarsBaR approach: 
>> http://marsbar.sourceforge.net/faq.html#how-is-the-percent-signal-change-calculated)
>>  
>> I'm also going to need the beta weights for the session mean 
>> regressors.  How are the nuisance regressors organized?
> You can just use the meanfunc.nii.gz. Also, each contrasts is 
> computed as the simple contrast (ces) and as a percent of the 
> baseline at the voxel (cespct, cesvarpct).
>

Re: [Freesurfer] beta weights from FS-Fast analysis

[Douglas N Greve]

On 05/31/2013 02:45 PM, Joseph Dien wrote:
>
> On May 31, 2013, at 12:09 PM, Douglas N Greve 
> mailto:gr...@nmr.mgh.harvard.edu>> wrote:
>
>>
>> On 05/30/2013 04:37 PM, Joseph Dien wrote:
>>> Just to make sure I'm doing this right, I'm going to summarize what
>>> I've taken away from your answers and to ask some new questions. In
>>> order to present the results, I need two things:
>>>
>>> 1) A set of histograms (with error bars) for each cluster figure to
>>> show the % signal change for each of the four contrasts of interest.
>>> The cache.th20.pos.y.ocn.dat file only gives it for the condition
>>> where the cluster was significant so I can't use that.
>>> So I could use mri_label2vol to convert cache.th20.neg.sig.ocn.annot
>>> from the group level analysis to generate a mask for each cluster of
>>> interest.
>>> Then I could extract the value of the voxels from each
>>> subject's cespct file for each contrast, average them across the
>>> cluster ROI, then average them across each subject, to generate the
>>> histogram?
>>> This would suffice to give me the %age signal change?
>>> I would be doing these computations in Matlab using MRIread.
>> I don't understand. If you don't have a cluster for a contrast, how are
>> you defining the cluster? From another contrast?
>>
>
> Well, what the reviewer told me to do is if I present a figure with a 
> significant cluster for one condition, I should use that as an ROI to 
> calculate the %age signal change for all four conditions and present 
> it as a bar chart as part of the figure.  I think she wanted to be 
> able to get a more qualitative sense of the data patterns.
If you have a cluster annotation (created by mri_glmfit-sim) then you 
can apply that to each contrast using mri_segstats with the --annot option.

>
>>>
>>> 2) A results table with the headings:
>>>
>>> Cluster p (FWE corrected)
>>> Cluster size
>>> Peak Voxel p (FWE corrected)
>>> Peak Voxel T
>>> Peak Voxel Coords
>>> BA
>>> Anatomical Landmark
>>>
>>> I can get the first two from
>>> the cache.th20.pos/neg.sig.cluster.summary files from the group level
>>> analysis.
>>> I can get the peak voxel coordinates from the summary files as well.
>>> I can use this to get the peak voxel p from the group level sig.nii.gz
>>> file.  Is this FWE corrected?  If not, how can I get this information?
>> What do you mean? The cluster p-value is corrected, why do you need the
>> max p and why does it need to be corrected?
>
> Well, as I understand it, the drawback of clusterwise statistics is 
> that while it assures you that the cluster passes muster as not being 
> due to random chance (at 95% confidence), it doesn't provide any 
> assurances at the voxel level (or in this case the vertex level) as it 
> is likely that a cluster is composed of both signal and noise and you 
> don't know which part is which.  So if a cluster covers both BA44 and 
> BA45 (for example), you can't be sure whether the activation involves 
> BA44, BA45, or both.  A voxelwise correction is more conservative but 
> if it provides significance, it does allow for this kind of 
> interpretation.
>
>>> I can use these coordinates to get the peak voxel T by getting the
>>> value from the group level F.nii.gz file and taking its square root.
>>> How can I get the sign of the T statistic?
>> Same as the sign of gamma.mgh
>
> Ah, great!
>
>>> I can use the Lancaster transform to convert the MNI305 peak voxel
>>> coordinates into the Atlas coordinates to look up the putative BA and
>>> landmarks (unless there is a better way with Freesurfer?  I'm seeing
>>> some references to some BA labels in the forum but it doesn't look
>>> like this is a complete set yet?).
>> Some of the BA labels are in FS, but not nearly all of them
>> doug
>
> No problem!  I worked out that I can use the talairach.nii file made 
> available by the Talairach Daemon folks.
>
>
>>>
>>> Sorry for all these questions!  I got some nice results from FSFAST
>>> and would like to get them written up.
>>>
>>> Cheers!
>>>
>>> Joe
>>>
>>>
>>>
>>>
>>> On May 29, 2013, at 10:53 PM, Douglas Greve 
>>> mailto:gr...@nmr.mgh.harvard.edu>
>>> > wrote:
>>>

 On 5/29/13 10:42 PM, Joseph Dien wrote:
>
> On May 29, 2013, at 11:40 AM, Douglas N Greve
> mailto:gr...@nmr.mgh.harvard.edu> 
> > wrote:
>
>> Hi Joe,
>>
>> On 05/29/2013 01:00 AM, Joseph Dien wrote:
>>> I need to extract the beta weights from a cluster identified with
>>> FS-Fast in order to compute percentage signal change.
>>>
>>> 1) I see a file called beta.nii.gz that appears to have the beta
>>> weight information.  It has a four dimensional structure and the
>>> fourth dimension appears to be the beta weights.  Is there an index
>>> somewhere as to which beta weight is which?  Or if not, how are they
>>> organized?
>> For the first level analysis, the first N beta weights correspond

Re: [Freesurfer] beta weights from FS-Fast analysis

[Joseph Dien]

On May 31, 2013, at 12:11 PM, Douglas N Greve  wrote:

> 
> On 05/31/2013 01:49 AM, Joseph Dien wrote:
>> I was able to make more progress so I'm mostly good at this point but 
>> I have a remaining question:
>> 
>> I assume the contents of sig.nii.gz (which I assume are the vertex 
>> p-values) are not FWE corrected.  Is it possible to get FWE-corrected 
>> vertex p-values?  Or are only clusterwise corrections available?
> There should be something like cache.th13.abs.sig.voxel.mgh which is 
> corrected on a voxelwise basis (the th13 is just part of the name but it 
> should be the same regardless of the threshold you choose)
> doug

Excellent!  Thanks!  :)

>> 
>> Thanks again for your patience!
>> 
>> Joe
>> 
>> On May 30, 2013, at 4:37 PM, Joseph Dien > > wrote:
>> 
>>> Just to make sure I'm doing this right, I'm going to summarize what 
>>> I've taken away from your answers and to ask some new questions. In 
>>> order to present the results, I need two things:
>>> 
>>> 1) A set of histograms (with error bars) for each cluster figure to 
>>> show the % signal change for each of the four contrasts of interest.
>>> The cache.th20.pos.y.ocn.dat file only gives it for the condition 
>>> where the cluster was significant so I can't use that.
>>> So I could use mri_label2vol to convert cache.th20.neg.sig.ocn.annot 
>>> from the group level analysis to generate a mask for each cluster of 
>>> interest.
>>> Then I could extract the value of the voxels from each 
>>> subject's cespct file for each contrast, average them across the 
>>> cluster ROI, then average them across each subject, to generate the 
>>> histogram?
>>> This would suffice to give me the %age signal change?
>>> I would be doing these computations in Matlab using MRIread.
>>> 
>>> 2) A results table with the headings:
>>> 
>>> Cluster p (FWE corrected)
>>> Cluster size
>>> Peak Voxel p (FWE corrected)
>>> Peak Voxel T
>>> Peak Voxel Coords
>>> BA
>>> Anatomical Landmark
>>> 
>>> I can get the first two from 
>>> the cache.th20.pos/neg.sig.cluster.summary files from the group level 
>>> analysis.
>>> I can get the peak voxel coordinates from the summary files as well.
>>> I can use this to get the peak voxel p from the group 
>>> level sig.nii.gz file.  Is this FWE corrected?  If not, how can I get 
>>> this information?
>>> I can use these coordinates to get the peak voxel T by getting the 
>>> value from the group level F.nii.gz file and taking its square root. 
>>> How can I get the sign of the T statistic?
>>> I can use the Lancaster transform to convert the MNI305 peak voxel 
>>> coordinates into the Atlas coordinates to look up the putative BA and 
>>> landmarks (unless there is a better way with Freesurfer?  I'm seeing 
>>> some references to some BA labels in the forum but it doesn't look 
>>> like this is a complete set yet?).
>>> 
>>> Sorry for all these questions!  I got some nice results from FSFAST 
>>> and would like to get them written up.
>>> 
>>> Cheers!
>>> 
>>> Joe
>>> 
>>> 
>>> 
>>> 
>>> On May 29, 2013, at 10:53 PM, Douglas Greve 
>>> mailto:gr...@nmr.mgh.harvard.edu>> wrote:
>>> 
 
 On 5/29/13 10:42 PM, Joseph Dien wrote:
> 
> On May 29, 2013, at 11:40 AM, Douglas N Greve 
> mailto:gr...@nmr.mgh.harvard.edu>> wrote:
> 
>> Hi Joe,
>> 
>> On 05/29/2013 01:00 AM, Joseph Dien wrote:
>>> I need to extract the beta weights from a cluster identified with
>>> FS-Fast in order to compute percentage signal change.
>>> 
>>> 1) I see a file called beta.nii.gz that appears to have the beta
>>> weight information.  It has a four dimensional structure and the
>>> fourth dimension appears to be the beta weights.  Is there an index
>>> somewhere as to which beta weight is which?  Or if not, how are they
>>> organized?
>> For the first level analysis, the first N beta weights correspond 
>> to the
>> N conditions in the paradigm file. The rest are nuisance variables.
>>> 
> 
> Ah, very good!  In order to compute the percent signal change 
> statistic (I'm following the MarsBaR approach: 
> http://marsbar.sourceforge.net/faq.html#how-is-the-percent-signal-change-calculated)
>  
> I'm also going to need the beta weights for the session mean 
> regressors.  How are the nuisance regressors organized?
 You can just use the meanfunc.nii.gz. Also, each contrasts is 
 computed as the simple contrast (ces) and as a percent of the 
 baseline at the voxel (cespct, cesvarpct).
> 
>>> 2) In order to extract the cluster, it looks like I would
>>> use mri_label2vol to convert cache.th20.neg.sig.ocn.annot into a
>>> volume where the voxels are tagged with the number of the
>>> corresponding cluster.
>> Is that  from a group analysis?
>>> 
> 
> Yes, that's right.
> 
>>> I could then use that to generate masks to extract the information I
>>> need for each cluster f

Re: [Freesurfer] beta weights from FS-Fast analysis

[Joseph Dien]

On May 31, 2013, at 12:09 PM, Douglas N Greve  wrote:

> 
> On 05/30/2013 04:37 PM, Joseph Dien wrote:
>> Just to make sure I'm doing this right, I'm going to summarize what 
>> I've taken away from your answers and to ask some new questions. In 
>> order to present the results, I need two things:
>> 
>> 1) A set of histograms (with error bars) for each cluster figure to 
>> show the % signal change for each of the four contrasts of interest.
>> The cache.th20.pos.y.ocn.dat file only gives it for the condition 
>> where the cluster was significant so I can't use that.
>> So I could use mri_label2vol to convert cache.th20.neg.sig.ocn.annot 
>> from the group level analysis to generate a mask for each cluster of 
>> interest.
>> Then I could extract the value of the voxels from each 
>> subject's cespct file for each contrast, average them across the 
>> cluster ROI, then average them across each subject, to generate the 
>> histogram?
>> This would suffice to give me the %age signal change?
>> I would be doing these computations in Matlab using MRIread.
> I don't understand. If you don't have a cluster for a contrast, how are 
> you defining the cluster? From another contrast?
> 

Well, what the reviewer told me to do is if I present a figure with a 
significant cluster for one condition, I should use that as an ROI to calculate 
the %age signal change for all four conditions and present it as a bar chart as 
part of the figure.  I think she wanted to be able to get a more qualitative 
sense of the data patterns.

>> 
>> 2) A results table with the headings:
>> 
>> Cluster p (FWE corrected)
>> Cluster size
>> Peak Voxel p (FWE corrected)
>> Peak Voxel T
>> Peak Voxel Coords
>> BA
>> Anatomical Landmark
>> 
>> I can get the first two from 
>> the cache.th20.pos/neg.sig.cluster.summary files from the group level 
>> analysis.
>> I can get the peak voxel coordinates from the summary files as well.
>> I can use this to get the peak voxel p from the group level sig.nii.gz 
>> file.  Is this FWE corrected?  If not, how can I get this information?
> What do you mean? The cluster p-value is corrected, why do you need the 
> max p and why does it need to be corrected?

Well, as I understand it, the drawback of clusterwise statistics is that while 
it assures you that the cluster passes muster as not being due to random chance 
(at 95% confidence), it doesn't provide any assurances at the voxel level (or 
in this case the vertex level) as it is likely that a cluster is composed of 
both signal and noise and you don't know which part is which.  So if a cluster 
covers both BA44 and BA45 (for example), you can't be sure whether the 
activation involves BA44, BA45, or both.  A voxelwise correction is more 
conservative but if it provides significance, it does allow for this kind of 
interpretation.

>> I can use these coordinates to get the peak voxel T by getting the 
>> value from the group level F.nii.gz file and taking its square root. 
>> How can I get the sign of the T statistic?
> Same as the sign of gamma.mgh

Ah, great!

>> I can use the Lancaster transform to convert the MNI305 peak voxel 
>> coordinates into the Atlas coordinates to look up the putative BA and 
>> landmarks (unless there is a better way with Freesurfer?  I'm seeing 
>> some references to some BA labels in the forum but it doesn't look 
>> like this is a complete set yet?).
> Some of the BA labels are in FS, but not nearly all of them
> doug

No problem!  I worked out that I can use the talairach.nii file made available 
by the Talairach Daemon folks.


>> 
>> Sorry for all these questions!  I got some nice results from FSFAST 
>> and would like to get them written up.
>> 
>> Cheers!
>> 
>> Joe
>> 
>> 
>> 
>> 
>> On May 29, 2013, at 10:53 PM, Douglas Greve > > wrote:
>> 
>>> 
>>> On 5/29/13 10:42 PM, Joseph Dien wrote:
 
 On May 29, 2013, at 11:40 AM, Douglas N Greve 
 mailto:gr...@nmr.mgh.harvard.edu>> wrote:
 
> Hi Joe,
> 
> On 05/29/2013 01:00 AM, Joseph Dien wrote:
>> I need to extract the beta weights from a cluster identified with
>> FS-Fast in order to compute percentage signal change.
>> 
>> 1) I see a file called beta.nii.gz that appears to have the beta
>> weight information.  It has a four dimensional structure and the
>> fourth dimension appears to be the beta weights.  Is there an index
>> somewhere as to which beta weight is which?  Or if not, how are they
>> organized?
> For the first level analysis, the first N beta weights correspond 
> to the
> N conditions in the paradigm file. The rest are nuisance variables.
>> 
 
 Ah, very good!  In order to compute the percent signal change 
 statistic (I'm following the MarsBaR approach: 
 http://marsbar.sourceforge.net/faq.html#how-is-the-percent-signal-change-calculated)
  
 I'm also going to need the beta weights for the session mean 
 regressors.

Re: [Freesurfer] beta weights from FS-Fast analysis

[Douglas N Greve]

On 05/31/2013 01:49 AM, Joseph Dien wrote:
> I was able to make more progress so I'm mostly good at this point but 
> I have a remaining question:
>
> I assume the contents of sig.nii.gz (which I assume are the vertex 
> p-values) are not FWE corrected.  Is it possible to get FWE-corrected 
> vertex p-values?  Or are only clusterwise corrections available?
There should be something like cache.th13.abs.sig.voxel.mgh which is 
corrected on a voxelwise basis (the th13 is just part of the name but it 
should be the same regardless of the threshold you choose)
doug
>
> Thanks again for your patience!
>
> Joe
>
> On May 30, 2013, at 4:37 PM, Joseph Dien  > wrote:
>
>> Just to make sure I'm doing this right, I'm going to summarize what 
>> I've taken away from your answers and to ask some new questions. In 
>> order to present the results, I need two things:
>>
>> 1) A set of histograms (with error bars) for each cluster figure to 
>> show the % signal change for each of the four contrasts of interest.
>> The cache.th20.pos.y.ocn.dat file only gives it for the condition 
>> where the cluster was significant so I can't use that.
>> So I could use mri_label2vol to convert cache.th20.neg.sig.ocn.annot 
>> from the group level analysis to generate a mask for each cluster of 
>> interest.
>> Then I could extract the value of the voxels from each 
>> subject's cespct file for each contrast, average them across the 
>> cluster ROI, then average them across each subject, to generate the 
>> histogram?
>> This would suffice to give me the %age signal change?
>> I would be doing these computations in Matlab using MRIread.
>>
>> 2) A results table with the headings:
>>
>> Cluster p (FWE corrected)
>> Cluster size
>> Peak Voxel p (FWE corrected)
>> Peak Voxel T
>> Peak Voxel Coords
>> BA
>> Anatomical Landmark
>>
>> I can get the first two from 
>> the cache.th20.pos/neg.sig.cluster.summary files from the group level 
>> analysis.
>> I can get the peak voxel coordinates from the summary files as well.
>> I can use this to get the peak voxel p from the group 
>> level sig.nii.gz file.  Is this FWE corrected?  If not, how can I get 
>> this information?
>> I can use these coordinates to get the peak voxel T by getting the 
>> value from the group level F.nii.gz file and taking its square root. 
>>  How can I get the sign of the T statistic?
>> I can use the Lancaster transform to convert the MNI305 peak voxel 
>> coordinates into the Atlas coordinates to look up the putative BA and 
>> landmarks (unless there is a better way with Freesurfer?  I'm seeing 
>> some references to some BA labels in the forum but it doesn't look 
>> like this is a complete set yet?).
>>
>> Sorry for all these questions!  I got some nice results from FSFAST 
>> and would like to get them written up.
>>
>> Cheers!
>>
>> Joe
>>
>>
>>
>>
>> On May 29, 2013, at 10:53 PM, Douglas Greve 
>> mailto:gr...@nmr.mgh.harvard.edu>> wrote:
>>
>>>
>>> On 5/29/13 10:42 PM, Joseph Dien wrote:

 On May 29, 2013, at 11:40 AM, Douglas N Greve 
 mailto:gr...@nmr.mgh.harvard.edu>> wrote:

> Hi Joe,
>
> On 05/29/2013 01:00 AM, Joseph Dien wrote:
>> I need to extract the beta weights from a cluster identified with
>> FS-Fast in order to compute percentage signal change.
>>
>> 1) I see a file called beta.nii.gz that appears to have the beta
>> weight information.  It has a four dimensional structure and the
>> fourth dimension appears to be the beta weights.  Is there an index
>> somewhere as to which beta weight is which?  Or if not, how are they
>> organized?
> For the first level analysis, the first N beta weights correspond 
> to the
> N conditions in the paradigm file. The rest are nuisance variables.
>>

 Ah, very good!  In order to compute the percent signal change 
 statistic (I'm following the MarsBaR approach: 
 http://marsbar.sourceforge.net/faq.html#how-is-the-percent-signal-change-calculated)
  
 I'm also going to need the beta weights for the session mean 
 regressors.  How are the nuisance regressors organized?
>>> You can just use the meanfunc.nii.gz. Also, each contrasts is 
>>> computed as the simple contrast (ces) and as a percent of the 
>>> baseline at the voxel (cespct, cesvarpct).

>> 2) In order to extract the cluster, it looks like I would
>> use mri_label2vol to convert cache.th20.neg.sig.ocn.annot into a
>> volume where the voxels are tagged with the number of the
>> corresponding cluster.
> Is that  from a group analysis?
>>

 Yes, that's right.

>> I could then use that to generate masks to extract the information I
>> need for each cluster from beta.nii.gz.
> If this is from a group analysis, then there should already be a file
> there (something.y.ocn.dat) that has a value for each subject in the
> rows and a value for each cluster in the columns.
>>

Re: [Freesurfer] beta weights from FS-Fast analysis

[Douglas N Greve]

On 05/30/2013 04:37 PM, Joseph Dien wrote:
> Just to make sure I'm doing this right, I'm going to summarize what 
> I've taken away from your answers and to ask some new questions. In 
> order to present the results, I need two things:
>
> 1) A set of histograms (with error bars) for each cluster figure to 
> show the % signal change for each of the four contrasts of interest.
> The cache.th20.pos.y.ocn.dat file only gives it for the condition 
> where the cluster was significant so I can't use that.
> So I could use mri_label2vol to convert cache.th20.neg.sig.ocn.annot 
> from the group level analysis to generate a mask for each cluster of 
> interest.
> Then I could extract the value of the voxels from each 
> subject's cespct file for each contrast, average them across the 
> cluster ROI, then average them across each subject, to generate the 
> histogram?
> This would suffice to give me the %age signal change?
> I would be doing these computations in Matlab using MRIread.
I don't understand. If you don't have a cluster for a contrast, how are 
you defining the cluster? From another contrast?

>
> 2) A results table with the headings:
>
> Cluster p (FWE corrected)
> Cluster size
> Peak Voxel p (FWE corrected)
> Peak Voxel T
> Peak Voxel Coords
> BA
> Anatomical Landmark
>
> I can get the first two from 
> the cache.th20.pos/neg.sig.cluster.summary files from the group level 
> analysis.
> I can get the peak voxel coordinates from the summary files as well.
> I can use this to get the peak voxel p from the group level sig.nii.gz 
> file.  Is this FWE corrected?  If not, how can I get this information?
What do you mean? The cluster p-value is corrected, why do you need the 
max p and why does it need to be corrected?
> I can use these coordinates to get the peak voxel T by getting the 
> value from the group level F.nii.gz file and taking its square root. 
>  How can I get the sign of the T statistic?
Same as the sign of gamma.mgh
> I can use the Lancaster transform to convert the MNI305 peak voxel 
> coordinates into the Atlas coordinates to look up the putative BA and 
> landmarks (unless there is a better way with Freesurfer?  I'm seeing 
> some references to some BA labels in the forum but it doesn't look 
> like this is a complete set yet?).
Some of the BA labels are in FS, but not nearly all of them
doug
>
> Sorry for all these questions!  I got some nice results from FSFAST 
> and would like to get them written up.
>
> Cheers!
>
> Joe
>
>
>
>
> On May 29, 2013, at 10:53 PM, Douglas Greve  > wrote:
>
>>
>> On 5/29/13 10:42 PM, Joseph Dien wrote:
>>>
>>> On May 29, 2013, at 11:40 AM, Douglas N Greve 
>>> mailto:gr...@nmr.mgh.harvard.edu>> wrote:
>>>
 Hi Joe,

 On 05/29/2013 01:00 AM, Joseph Dien wrote:
> I need to extract the beta weights from a cluster identified with
> FS-Fast in order to compute percentage signal change.
>
> 1) I see a file called beta.nii.gz that appears to have the beta
> weight information.  It has a four dimensional structure and the
> fourth dimension appears to be the beta weights.  Is there an index
> somewhere as to which beta weight is which?  Or if not, how are they
> organized?
 For the first level analysis, the first N beta weights correspond 
 to the
 N conditions in the paradigm file. The rest are nuisance variables.
>
>>>
>>> Ah, very good!  In order to compute the percent signal change 
>>> statistic (I'm following the MarsBaR approach: 
>>> http://marsbar.sourceforge.net/faq.html#how-is-the-percent-signal-change-calculated)
>>>  
>>> I'm also going to need the beta weights for the session mean 
>>> regressors.  How are the nuisance regressors organized?
>> You can just use the meanfunc.nii.gz. Also, each contrasts is 
>> computed as the simple contrast (ces) and as a percent of the 
>> baseline at the voxel (cespct, cesvarpct).
>>>
> 2) In order to extract the cluster, it looks like I would
> use mri_label2vol to convert cache.th20.neg.sig.ocn.annot into a
> volume where the voxels are tagged with the number of the
> corresponding cluster.
 Is that  from a group analysis?
>
>>>
>>> Yes, that's right.
>>>
> I could then use that to generate masks to extract the information I
> need for each cluster from beta.nii.gz.
 If this is from a group analysis, then there should already be a file
 there (something.y.ocn.dat) that has a value for each subject in the
 rows and a value for each cluster in the columns.
>
>>>
>>> I see it.  Are these values already scaled as percent signal change? 
>>>  If so, that would be wonderful!  :)
>> Only if you specified it when you ran isxconcat-sess. Note that the 
>> "non-scaled" values are actually scaled to percent of grand mean 
>> intensity.
>>>
> Is that correct?
>
> 3) The final information that I would need is the canonical hrf shape
> generated by FSFAST for a single even

Re: [Freesurfer] beta weights from FS-Fast analysis

[Joseph Dien]

I was able to make more progress so I'm mostly good at this point but I have a 
remaining question:

I assume the contents of sig.nii.gz (which I assume are the vertex p-values) 
are not FWE corrected.  Is it possible to get FWE-corrected vertex p-values?  
Or are only clusterwise corrections available?

Thanks again for your patience!

Joe

On May 30, 2013, at 4:37 PM, Joseph Dien  wrote:

> Just to make sure I'm doing this right, I'm going to summarize what I've 
> taken away from your answers and to ask some new questions. In order to 
> present the results, I need two things:
> 
> 1) A set of histograms (with error bars) for each cluster figure to show the 
> % signal change for each of the four contrasts of interest.
> The cache.th20.pos.y.ocn.dat file only gives it for the condition where the 
> cluster was significant so I can't use that.
> So I could use mri_label2vol to convert cache.th20.neg.sig.ocn.annot from the 
> group level analysis to generate a mask for each cluster of interest.
> Then I could extract the value of the voxels from each subject's cespct file 
> for each contrast, average them across the cluster ROI, then average them 
> across each subject, to generate the histogram?
> This would suffice to give me the %age signal change?
> I would be doing these computations in Matlab using MRIread.
> 
> 2) A results table with the headings: 
> 
> Cluster p (FWE corrected)
> Cluster size
> Peak Voxel p (FWE corrected)
> Peak Voxel T
> Peak Voxel Coords
> BA
> Anatomical Landmark
> 
> I can get the first two from the cache.th20.pos/neg.sig.cluster.summary files 
> from the group level analysis.
> I can get the peak voxel coordinates from the summary files as well.
> I can use this to get the peak voxel p from the group level sig.nii.gz file.  
> Is this FWE corrected?  If not, how can I get this information?
> I can use these coordinates to get the peak voxel T by getting the value from 
> the group level F.nii.gz file and taking its square root.  How can I get the 
> sign of the T statistic?
> I can use the Lancaster transform to convert the MNI305 peak voxel 
> coordinates into the Atlas coordinates to look up the putative BA and 
> landmarks (unless there is a better way with Freesurfer?  I'm seeing some 
> references to some BA labels in the forum but it doesn't look like this is a 
> complete set yet?).
> 
> Sorry for all these questions!  I got some nice results from FSFAST and would 
> like to get them written up.
> 
> Cheers!
> 
> Joe
> 
> 
> 
> 
> On May 29, 2013, at 10:53 PM, Douglas Greve  wrote:
> 
>> 
>> On 5/29/13 10:42 PM, Joseph Dien wrote:
>>> 
>>> On May 29, 2013, at 11:40 AM, Douglas N Greve  
>>> wrote:
>>> 
 Hi Joe,
 
 On 05/29/2013 01:00 AM, Joseph Dien wrote:
> I need to extract the beta weights from a cluster identified with 
> FS-Fast in order to compute percentage signal change.
> 
> 1) I see a file called beta.nii.gz that appears to have the beta 
> weight information.  It has a four dimensional structure and the 
> fourth dimension appears to be the beta weights.  Is there an index 
> somewhere as to which beta weight is which?  Or if not, how are they 
> organized?
 For the first level analysis, the first N beta weights correspond to the 
 N conditions in the paradigm file. The rest are nuisance variables.
> 
>>> 
>>> Ah, very good!  In order to compute the percent signal change statistic 
>>> (I'm following the MarsBaR approach: 
>>> http://marsbar.sourceforge.net/faq.html#how-is-the-percent-signal-change-calculated)
>>>  I'm also going to need the beta weights for the session mean regressors.  
>>> How are the nuisance regressors organized?
>> You can just use the meanfunc.nii.gz. Also, each contrasts is computed as 
>> the simple contrast (ces) and as a percent of the baseline at the voxel 
>> (cespct, cesvarpct).
>>> 
> 2) In order to extract the cluster, it looks like I would 
> use mri_label2vol to convert cache.th20.neg.sig.ocn.annot into a 
> volume where the voxels are tagged with the number of the 
> corresponding cluster.
 Is that  from a group analysis?
> 
>>> 
>>> Yes, that's right.
>>> 
> I could then use that to generate masks to extract the information I 
> need for each cluster from beta.nii.gz.
 If this is from a group analysis, then there should already be a file 
 there (something.y.ocn.dat) that has a value for each subject in the 
 rows and a value for each cluster in the columns.
> 
>>> 
>>> I see it.  Are these values already scaled as percent signal change?  If 
>>> so, that would be wonderful!  :)
>> Only if you specified it when you ran isxconcat-sess. Note that the 
>> "non-scaled" values are actually scaled to percent of grand mean intensity.
>>> 
> Is that correct?
> 
> 3) The final information that I would need is the canonical hrf shape 
> generated by FSFAST for a single event.  I guess I could generate that 
>>>

Re: [Freesurfer] beta weights from FS-Fast analysis

[Joseph Dien]

Just to make sure I'm doing this right, I'm going to summarize what I've taken 
away from your answers and to ask some new questions. In order to present the 
results, I need two things:

1) A set of histograms (with error bars) for each cluster figure to show the % 
signal change for each of the four contrasts of interest.
The cache.th20.pos.y.ocn.dat file only gives it for the condition where the 
cluster was significant so I can't use that.
So I could use mri_label2vol to convert cache.th20.neg.sig.ocn.annot from the 
group level analysis to generate a mask for each cluster of interest.
Then I could extract the value of the voxels from each subject's cespct file 
for each contrast, average them across the cluster ROI, then average them 
across each subject, to generate the histogram?
This would suffice to give me the %age signal change?
I would be doing these computations in Matlab using MRIread.

2) A results table with the headings: 

Cluster p (FWE corrected)
Cluster size
Peak Voxel p (FWE corrected)
Peak Voxel T
Peak Voxel Coords
BA
Anatomical Landmark

I can get the first two from the cache.th20.pos/neg.sig.cluster.summary files 
from the group level analysis.
I can get the peak voxel coordinates from the summary files as well.
I can use this to get the peak voxel p from the group level sig.nii.gz file.  
Is this FWE corrected?  If not, how can I get this information?
I can use these coordinates to get the peak voxel T by getting the value from 
the group level F.nii.gz file and taking its square root.  How can I get the 
sign of the T statistic?
I can use the Lancaster transform to convert the MNI305 peak voxel coordinates 
into the Atlas coordinates to look up the putative BA and landmarks (unless 
there is a better way with Freesurfer?  I'm seeing some references to some BA 
labels in the forum but it doesn't look like this is a complete set yet?).

Sorry for all these questions!  I got some nice results from FSFAST and would 
like to get them written up.

Cheers!

Joe




On May 29, 2013, at 10:53 PM, Douglas Greve  wrote:

> 
> On 5/29/13 10:42 PM, Joseph Dien wrote:
>> 
>> On May 29, 2013, at 11:40 AM, Douglas N Greve  
>> wrote:
>> 
>>> Hi Joe,
>>> 
>>> On 05/29/2013 01:00 AM, Joseph Dien wrote:
 I need to extract the beta weights from a cluster identified with 
 FS-Fast in order to compute percentage signal change.
 
 1) I see a file called beta.nii.gz that appears to have the beta 
 weight information.  It has a four dimensional structure and the 
 fourth dimension appears to be the beta weights.  Is there an index 
 somewhere as to which beta weight is which?  Or if not, how are they 
 organized?
>>> For the first level analysis, the first N beta weights correspond to the 
>>> N conditions in the paradigm file. The rest are nuisance variables.
 
>> 
>> Ah, very good!  In order to compute the percent signal change statistic (I'm 
>> following the MarsBaR approach: 
>> http://marsbar.sourceforge.net/faq.html#how-is-the-percent-signal-change-calculated)
>>  I'm also going to need the beta weights for the session mean regressors.  
>> How are the nuisance regressors organized?
> You can just use the meanfunc.nii.gz. Also, each contrasts is computed as the 
> simple contrast (ces) and as a percent of the baseline at the voxel (cespct, 
> cesvarpct).
>> 
 2) In order to extract the cluster, it looks like I would 
 use mri_label2vol to convert cache.th20.neg.sig.ocn.annot into a 
 volume where the voxels are tagged with the number of the 
 corresponding cluster.
>>> Is that  from a group analysis?
 
>> 
>> Yes, that's right.
>> 
 I could then use that to generate masks to extract the information I 
 need for each cluster from beta.nii.gz.
>>> If this is from a group analysis, then there should already be a file 
>>> there (something.y.ocn.dat) that has a value for each subject in the 
>>> rows and a value for each cluster in the columns.
 
>> 
>> I see it.  Are these values already scaled as percent signal change?  If so, 
>> that would be wonderful!  :)
> Only if you specified it when you ran isxconcat-sess. Note that the 
> "non-scaled" values are actually scaled to percent of grand mean intensity.
>> 
 Is that correct?
 
 3) The final information that I would need is the canonical hrf shape 
 generated by FSFAST for a single event.  I guess I could generate that 
 by setting up a dummy analysis run with a single event of the desired 
 duration and then look in the X variable in the resulting X.mat file?
>>> try this
>>> plot(X.runflac(1).flac.ev(2).tirf, X.runflac(1).flac.ev(2).Xirf)
 
>> 
>> Perfect!  :)
>> 
 Sorry for all the questions!
 
 Joe
 
 
 
 
 
 
 Joseph Dien,
 Senior Research Scientist
 University of Maryland
 
 E-mail: jdie...@mac.com 

Re: [Freesurfer] beta weights from FS-Fast analysis

[Douglas Greve]

On 5/29/13 10:42 PM, Joseph Dien wrote:


On May 29, 2013, at 11:40 AM, Douglas N Greve 
mailto:gr...@nmr.mgh.harvard.edu>> wrote:



Hi Joe,

On 05/29/2013 01:00 AM, Joseph Dien wrote:

I need to extract the beta weights from a cluster identified with
FS-Fast in order to compute percentage signal change.

1) I see a file called beta.nii.gz that appears to have the beta
weight information.  It has a four dimensional structure and the
fourth dimension appears to be the beta weights.  Is there an index
somewhere as to which beta weight is which?  Or if not, how are they
organized?

For the first level analysis, the first N beta weights correspond to the
N conditions in the paradigm file. The rest are nuisance variables.




Ah, very good!  In order to compute the percent signal change 
statistic (I'm following the MarsBaR approach: 
http://marsbar.sourceforge.net/faq.html#how-is-the-percent-signal-change-calculated) 
I'm also going to need the beta weights for the session mean 
regressors.  How are the nuisance regressors organized?
You can just use the meanfunc.nii.gz. Also, each contrasts is computed 
as the simple contrast (ces) and as a percent of the baseline at the 
voxel (cespct, cesvarpct).



2) In order to extract the cluster, it looks like I would
use mri_label2vol to convert cache.th20.neg.sig.ocn.annot into a
volume where the voxels are tagged with the number of the
corresponding cluster.

Is that  from a group analysis?




Yes, that's right.


I could then use that to generate masks to extract the information I
need for each cluster from beta.nii.gz.

If this is from a group analysis, then there should already be a file
there (something.y.ocn.dat) that has a value for each subject in the
rows and a value for each cluster in the columns.




I see it.  Are these values already scaled as percent signal change? 
 If so, that would be wonderful!  :)
Only if you specified it when you ran isxconcat-sess. Note that the 
"non-scaled" values are actually scaled to percent of grand mean intensity.



Is that correct?

3) The final information that I would need is the canonical hrf shape
generated by FSFAST for a single event.  I guess I could generate that
by setting up a dummy analysis run with a single event of the desired
duration and then look in the X variable in the resulting X.mat file?

try this
plot(X.runflac(1).flac.ev(2).tirf, X.runflac(1).flac.ev(2).Xirf)




Perfect!  :)


Sorry for all the questions!

Joe






Joseph Dien,
Senior Research Scientist
University of Maryland

E-mail: jdie...@mac.com  


Phone: 301-226-8848
Fax: 301-226-8811
http://joedien.com//













___
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: https://gate.nmr.mgh.harvard.edu/filedrop2
www.nmr.mgh.harvard.edu/facility/filedrop/index.html
Outgoing: ftp://surfer.nmr.mgh.harvard.edu/transfer/outgoing/flat/greve/

___
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.






Joseph Dien,
Senior Research Scientist
University of Maryland

E-mail: jdie...@mac.com 
Phone: 301-226-8848
Fax: 301-226-8811
http://joedien.com//













___
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.

Re: [Freesurfer] beta weights from FS-Fast analysis

[Joseph Dien]

On May 29, 2013, at 11:40 AM, Douglas N Greve  wrote:

> Hi Joe,
> 
> On 05/29/2013 01:00 AM, Joseph Dien wrote:
>> I need to extract the beta weights from a cluster identified with 
>> FS-Fast in order to compute percentage signal change.
>> 
>> 1) I see a file called beta.nii.gz that appears to have the beta 
>> weight information.  It has a four dimensional structure and the 
>> fourth dimension appears to be the beta weights.  Is there an index 
>> somewhere as to which beta weight is which?  Or if not, how are they 
>> organized?
> For the first level analysis, the first N beta weights correspond to the 
> N conditions in the paradigm file. The rest are nuisance variables.
>> 

Ah, very good!  In order to compute the percent signal change statistic (I'm 
following the MarsBaR approach: 
http://marsbar.sourceforge.net/faq.html#how-is-the-percent-signal-change-calculated)
 I'm also going to need the beta weights for the session mean regressors.  How 
are the nuisance regressors organized?

>> 2) In order to extract the cluster, it looks like I would 
>> use mri_label2vol to convert cache.th20.neg.sig.ocn.annot into a 
>> volume where the voxels are tagged with the number of the 
>> corresponding cluster.
> Is that  from a group analysis?
>> 

Yes, that's right.

>> I could then use that to generate masks to extract the information I 
>> need for each cluster from beta.nii.gz.
> If this is from a group analysis, then there should already be a file 
> there (something.y.ocn.dat) that has a value for each subject in the 
> rows and a value for each cluster in the columns.
>> 

I see it.  Are these values already scaled as percent signal change?  If so, 
that would be wonderful!  :)

>> Is that correct?
>> 
>> 3) The final information that I would need is the canonical hrf shape 
>> generated by FSFAST for a single event.  I guess I could generate that 
>> by setting up a dummy analysis run with a single event of the desired 
>> duration and then look in the X variable in the resulting X.mat file?
> try this
> plot(X.runflac(1).flac.ev(2).tirf, X.runflac(1).flac.ev(2).Xirf)
>> 

Perfect!  :)

>> Sorry for all the questions!
>> 
>> Joe
>> 
>> 
>> 
>> 
>> 
>> 
>> Joseph Dien,
>> Senior Research Scientist
>> University of Maryland
>> 
>> E-mail: jdie...@mac.com 
>> Phone: 301-226-8848
>> Fax: 301-226-8811
>> http://joedien.com//
>> 
>> 
>> 
>> 
>> 
>> 
>> 
>> 
>> 
>> 
>> 
>> 
>> 
>> ___
>> 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: https://gate.nmr.mgh.harvard.edu/filedrop2
> www.nmr.mgh.harvard.edu/facility/filedrop/index.html
> Outgoing: ftp://surfer.nmr.mgh.harvard.edu/transfer/outgoing/flat/greve/
> 
> ___
> 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.
> 




Joseph Dien,
Senior Research Scientist
University of Maryland 

E-mail: jdie...@mac.com
Phone: 301-226-8848
Fax: 301-226-8811
http://joedien.com//











___
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.

Re: [Freesurfer] beta weights from FS-Fast analysis

[Douglas N Greve]

Hi Joe,

On 05/29/2013 01:00 AM, Joseph Dien wrote:
> I need to extract the beta weights from a cluster identified with 
> FS-Fast in order to compute percentage signal change.
>
> 1) I see a file called beta.nii.gz that appears to have the beta 
> weight information.  It has a four dimensional structure and the 
> fourth dimension appears to be the beta weights.  Is there an index 
> somewhere as to which beta weight is which?  Or if not, how are they 
> organized?
For the first level analysis, the first N beta weights correspond to the 
N conditions in the paradigm file. The rest are nuisance variables.
>
> 2) In order to extract the cluster, it looks like I would 
> use mri_label2vol to convert cache.th20.neg.sig.ocn.annot into a 
> volume where the voxels are tagged with the number of the 
> corresponding cluster.
Is that  from a group analysis?
>
> I could then use that to generate masks to extract the information I 
> need for each cluster from beta.nii.gz.
If this is from a group analysis, then there should already be a file 
there (something.y.ocn.dat) that has a value for each subject in the 
rows and a value for each cluster in the columns.
>
> Is that correct?
>
> 3) The final information that I would need is the canonical hrf shape 
> generated by FSFAST for a single event.  I guess I could generate that 
> by setting up a dummy analysis run with a single event of the desired 
> duration and then look in the X variable in the resulting X.mat file?
try this
plot(X.runflac(1).flac.ev(2).tirf, X.runflac(1).flac.ev(2).Xirf)
>
> Sorry for all the questions!
>
> Joe
>
>
>
>
> 
>
> Joseph Dien,
> Senior Research Scientist
> University of Maryland
>
> E-mail: jdie...@mac.com 
> Phone: 301-226-8848
> Fax: 301-226-8811
> http://joedien.com//
>
>
>
>
>
>
>
>
>
>
>
>
>
> ___
> 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: https://gate.nmr.mgh.harvard.edu/filedrop2
www.nmr.mgh.harvard.edu/facility/filedrop/index.html
Outgoing: ftp://surfer.nmr.mgh.harvard.edu/transfer/outgoing/flat/greve/

___
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] beta weights from FS-Fast analysis

[Joseph Dien]

I need to extract the beta weights from a cluster identified with FS-Fast in 
order to compute percentage signal change.  

1) I see a file called beta.nii.gz that appears to have the beta weight 
information.  It has a four dimensional structure and the fourth dimension 
appears to be the beta weights.  Is there an index somewhere as to which beta 
weight is which?  Or if not, how are they organized?

2) In order to extract the cluster, it looks like I would use mri_label2vol to 
convert cache.th20.neg.sig.ocn.annot into a volume where the voxels are tagged 
with the number of the corresponding cluster.

I could then use that to generate masks to extract the information I need for 
each cluster from beta.nii.gz.

Is that correct?

3) The final information that I would need is the canonical hrf shape generated 
by FSFAST for a single event.  I guess I could generate that by setting up a 
dummy analysis run with a single event of the desired duration and then look in 
the X variable in the resulting X.mat file?

Sorry for all the questions!

Joe






Joseph Dien,
Senior Research Scientist
University of Maryland 

E-mail: jdie...@mac.com
Phone: 301-226-8848
Fax: 301-226-8811
http://joedien.com//











___
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.