Re: [Freesurfer] regional surface area differences

[Anderson M. Winkler]

Hi Abid,

What troubles did you have with Spherical Demons? In any case, though using
a different interpolation method, if the registration is the same,
mris_preproc gives results very similar as the pycnophylactic method we
described in the paper, as long as the data is smoothed. However, before
smoothing, you need to consider the different face sizes (in this case, the
"vertex sizes"), as we show in the Appendix A of the paper. Currently this
step would need to be done outside mris_preproc.

A suggestion for running is (there are other ways, though):

1) Run mris_preproc as:

mris_preproc --s ${subj1} --s ${subj2} --s ${subj3} [...] --target
fsaverage --hemi ${h} --meas area --out ${h}-area-mrispreproc.mgh
--nocleanup

2) The --nocleanup option will keep the files for each subject. These can
be converted to ASCII with mris_convert (-c), and then the outputs
subjected to the correction we discuss in the Appendix. This correction can
be done with the command "rpncalc" as you may have seen in the webpage with
the documentation for the pycnophylactic method.

3) Once that has been done, the data can be smoothed, and then converted
back to other formats so that you can run the statistical analysis via,
e.g., mris_glmfit.

Note the above won't be using SD registration.

Hope this helps!

All the best,

Anderson



On 27 February 2015 at 20:41,  wrote:

> Hi Freesurfer group,
>
> I wanted to use mris_preproc to calculate regional differences in surface
> area between two groups.
>
> Ideally I would like to use something that is vertex-wise and preserves
> regional area, but I've had trouble with spherical demons (using the
> Winkler methodology). So I am going to first try using mris_preproc (and
> compare the two methods). I found this old thread suggesting that I should
> use the --area flag.
> But I had some questions:
>
> 1) Which surfname should I use with the --area flag? area.mid vs area.pial
> vs any other file
> 2) Should I use --meas flag in addition and if so which one? (similar to
> how --meas thickness is used for calculating cortical thickness)
> 3) Which --surfreg file should I use? sphere.reg vs sphere. Or do I not
> need the --surfreg?
>
>
> Thanks,
> Abid
>
>
> -- Forwarded message --
> From: Anderson M. Winkler 
> Date: Tue, May 14, 2013 at 1:19 AM
> Subject: Re: [Freesurfer] Usage of --area option in estimating the
> vertex-wise maps of surface area
> To: free 
>
>
> Hi Xi-Nian,
>
> As Doug said, they don't do the same thing. The method in the paper
> uses a different kind of interpolation to allows the preservation of
> the amount of area, whereas mris_preproc achieves the same globally
> using a Jacobian correction. The results aren't expected to be
> identical at the vertex level or for small regions.
>
> An implementation of the method we proposed in the paper is available
> for Octave and/or Matlab at http://brainder.org/download/areal
> It works fine, but beware that it's quite slow.
>
> All the best,
>
> Anderson
>
>
>
> 2013/5/13 Douglas Greve 
> >
> >
> > Hi Xi-Nian, it does not do the same thing, but it does something
> similar. Anderson found that it gave similar results in a lot of cases,
> but he has found some differences. If he's still following the FS list,
> maybe he can comment.
> >
> > doug
> >
> >
> >
> >
> >
> > On 5/11/13 9:53 PM, Xinian Zuo wrote:
> >
> > Hello Doug,
> >
> > I am doing some association studies between functional measures and
> surface area. In using  mris_preproc, there is an option --area. Does
> this do sth similar to that as Winkler et al. (2012)?
> >
> > Winkler et al., 2012. Measureing and comparing brain cortical
> surfacearea and other areal quantities.
> >
> > Thank you very much.
>
> ___
> 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 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] volume, surface area, and thickness

[Anderson M. Winkler]

Hi Woo-Suk

Just adding to Bruce's reply: area isn't a "technically more noisy and
usually thus less sensitive" as the reviewer suggests. Area isn't noisier
on its own right, and it's measured from the very same surfaces from which
thickness is measured. However, there is a much larger variability of area
across subjects than of thickness, even within the normal range, such that
the variance of volume, that can be explained by or associated with other
indices, is largely explained by the variance in area. The first paper that
(as far as I know) showed this is Voets et al. (Neuroimage, 2008), and we
keep observing this repeatedly in different datasets, published or not.

About the Schmaal et al. paper (Molecular Psychiatry, 2016): it is an
excellent paper in which the authors didn't spend time (or space)
discussing volume, going instead straight to the more interesting bits:
area and thickness.

All the best,

Anderson


On 29 August 2016 at 13:56, Bruce Fischl  wrote:

> Hi Woo-Suk
>
> why not just do the surface analysis that they are requesting? I'm not
> sure what you are asking, but certainly volume = surface area * thickness
> in general, and so a volumetric effect can be driven by one or both of
> surface area and thickness
>
> cheers
> Bruce
>
>
>
> On Sat, 27 Aug 2016, Woo-Suk Tae wrote:
>
> Dear FreeSurfer experts and developers
>>
>> I am confronted with some technical question of a reviewer.
>> I am not sure about "volume = thickness-by-surface area" and the review's
>> opinion (attached below) about volume, surface area, and thickness was
>> correct.
>> Any comments from FreeSurfer's experts would help me.
>>
>> Sincerely yours
>>
>> Woo-Suk Tae
>> Seoul, Korea
>>
>> I added the reviewer's comments.
>> -
>> 1. Volume/thickness: The authors cite many papers that show volume and
>> thickness differences in MDD. The unresolved part here, however, is the
>> RELATION between cortical thickness and cortical volume: There is no
>> doubt,
>> that both measures are found affected in MDD. This is, because cortical
>> thickess multiplied by the surface area of a gyrus results in its volume,
>> so
>> volumse = thickness-by-surface area. Surface area values themselves are
>> technically more noisy and usually thus less sensitive (due to the problem
>> of false attributions to an area).
>>
>> So, volume is influenced by thickness and surface and is the less specific
>> measure. If a volume effect is detected in a study, it is unclear, if it
>> is
>> driven by thickness, or surface area, or both. In this respect, the study
>> of
>> Schmaal et al. is telling, as it analyzed BOTH measures, finding only
>> thickness effects of MDD, and (practically) no surface area changes except
>> for adolescent MDD. In the adolescent MDD samples gross surface area
>> differences were detected.
>>
>> This means, that the question of "superiority" is rather a question of
>> "specificity": (Cortical) volume findings in adult MDD are mostly driven
>> by
>> thickness differences and are in on way independent from thickness
>> differences. In this respect, the authors should follow the basic
>> geometric
>> principles of morphometry and point out the relatedness of the two. They
>> can
>> simply check this in their FreeSurfer variables (e. g. for total grey
>> matter
>> volume). Not reporting surface area is leaving an interpretational gap as
>> surface area differences could in addition drive volume differences. The
>> authors may want to decide not to present surface area results, but then
>> they should discuss this as limitation to disentangle the origin of
>> volumetric (cortical) effects. This seems important as methylation effects
>> and FKBP5 interact with early life time stress, so effects on surface area
>> as seen in adolescent MDD highlight that surface area effects could play
>> in.
>>
>>
>>
>>
>>
>>
>>
>> 
>> ---
>> -
>> Woo-Suk, Tae  Ph.D.  Research Professor
>> Brain Convergence Research Center, Medical Research Center
>> Anam Hospital, Korea University Medical Center, Seoul, Korea
>> mobile: 82-10-9120-4629
>> office: 82-2-920-6831
>> email: woosuk@gmail.com, woos...@gmail.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] mri_glmfit-sim permutation testing running after 3 days!

[Anderson M. Winkler]

Hi all,

In the presence of nuisance (orthogonal or not) the permutation test is
only approximate, whatever is the method used. Regressing out the nuisance
introduces dependencies between the residuals that render them strictly not
exchangeable. However, in practice this is not an issue, and the FL method
has been assessed extensively by us and also by others (e.g., Anderson &
Legendre 1999; Anderson & Robinson, 2001; Anderson & ter Braak, 2003; among
others) and does lead to error rates that approach the test level.

In the Winkler et al 2014 paper, Table 7 gives a sense of how good or bad
some of the regression and permutation methods can be, and even the FL
method, that we advocate, can lead to error rates below or above the 95%
CI, although the chances of invalid results are very small. This is for a
single test. When we consider the thousands of voxels/vertices, this gets
further diluted in the distribution of the maximum, such that the FWER is
even closer to perfectly exact (this information isn't in the paper,
though).

In the presence of non-orthogonality between regressors of interest and
nuisance, the result remains approximately exact, only the power to detect
a true effect is reduced, as it cannot be disambiguated from the nuisance.

Cheers,

Anderson


On 30 August 2016 at 09:33, Douglas Greve  wrote:

> By "wrong" I meant that permutation no longer  gives exact p-values in
> expectation with non-orthogonal designs. There is no theory to
> characterize the accuracy of  Freeman-Lane of the other methods. In this
> sense, they are not approximations but ad hoc methods that people hope
> do a better job than parametric methods. Anderson tested them on a wide
> range of designs, but it was, of course, not exhaustive. The accuracy of
> his results may not extend to other designs, so it is a buyer-beware
> situation (as with all neuroimaging).
>
>
> On 8/29/16 10:16 PM, Harms, Michael wrote:
> > Hi,
> > I wouldn’t say that non-orthogonal designs are “wrong” to use with
> > permutation.  Rather, there are different approaches to handling that
> > situation and produce approximate p-values.  See Table 2 in Winkler’s
> 2014
> > paper, and the results therein comparing the various approaches:
> >
> > http://www.ncbi.nlm.nih.gov/pubmed/24530839
> >
> >
> > PALM actually gives you control over the method used, with the default
> > (and recommended) approach being that of “Freedman-Lane", which is the
> > same approach used by FSL’s ‘randomise’ tool to handle correlated
> > covariates.
> >
> > cheers,
> > -MH
> >
> > --
> > Michael Harms, Ph.D.
> >
> > ---
> > Conte Center for the Neuroscience of Mental Disorders
> > Washington University School of Medicine
> > Department of Psychiatry, Box 8134
> > 660 South Euclid Ave.Tel: 314-747-6173
> > St. Louis, MO  63110Email: mha...@wustl.edu
> >
> >
> >
> >
> > On 8/29/16, 7:49 PM, "freesurfer-boun...@nmr.mgh.harvard.edu on behalf
> of
> > Matt Glasser"  > m...@ma-tea.com> wrote:
> >
> > PALM handles GIFTI and CIFTI data.
> >
> > Peace,
> >
> > Matt.
> >
> > On 8/29/16, 6:21 PM, "Douglas N Greve"
> >  > gr...@nmr.mgh.harvard.edu> wrote:
> >
> >> Does PALM do surface-based? Also, there is no way to appropriately
> >> handle this. For permutation, non-orthogonal designs are wrong. There
> >> are ways to try to compensate for it, which is what PALM is doing. Sorry
> >> to be nit-picky!
> >>
> >>
> >> On 08/29/2016 06:12 PM, Harms, Michael wrote:
> >>> Hi Maaike,
> >>> Why not just use PALM?  Then you don¹t have to worry about this (since
> >>> PALM appropriately handles the situation of correlated covariates).
> >>>
> >>> cheers,
> >>> -MH
> >>>
> >>> --
> >>> Michael Harms, Ph.D.
> >>>
> >>> ---
> >>> Conte Center for the Neuroscience of Mental Disorders
> >>> Washington University School of Medicine
> >>> Department of Psychiatry, Box 8134
> >>> 660 South Euclid Ave.Tel: 314-747-6173
> >>> St. Louis, MO  63110Email: mha...@wustl.edu
> >>>
> >>>
> >>>
> >>>
> >>> On 8/29/16, 4:45 PM, "freesurfer-boun...@nmr.mgh.harvard.edu on behalf
> >>> of
> >>> Douglas N Greve"  >>> gr...@nmr.mgh.harvard.edu> wrote:
> >>>
> >>> It is hard to say. Since the subjects are not exchangeable, the
> >>> permutation is technically not appropriate. Check the winkler paper,  I
> >>> think he talks about what happens if you just don't do anything.
> >>>
> >>>
> >>> On 08/29/2016 11:07 AM, maaike rive wrote:
>  Hi all,
> 
> 
>  Is using forced permutation for non-orthogonal design matrices wrong
>  or is it allowed to do this instead of using tools like palm (what
>  happens eg with the covariates when using forced permutation)? I
>  used forced permutation  and it seemed to work, results were (partly)
>  comparable to what I found with monte carlo simulations.
> 
> 
>  Thanks, Maaike
> 
> 
>  

Re: [Freesurfer] 3 fixed factors and matrix design

[Anderson M. Winkler]

Hi Shady,

Try assembling the design as this:

EV1: Group 1 (coded as 0/1)
EV2: Group 2 (coded as 0/1)
EV3: Group 3 (coded as 0/1)
EV4: Sex (coded as +1/-1)
EV5: Site (coded as +1/-1)

The contrasts to compare the three groups are then:
C1: [1 -1 0 0 0]
C2: [1 0 -1 0 0]
C3: [0 1 -1 0 0]
C4: [-1 1 0 0 0]
C5: [-1 0 1 0 0]
C6: [0 -1 1 0 0]

>From the description it seems site and group are confounded, although not
completely, such that it should be possible to still draw some inferences,
although without much power.

The above does not include terms for interactions. If it's something you
could be interested in, then extra EVs can be added.

All the best,

Anderson


On 30 August 2016 at 11:40, Shady Rahayel  wrote:

> Greetings,
>
> I want to design a matrix that contains 3 fixed factors. The first is
> diagnosis (3 levels), the second is gender (2 levels), and the last is
> imaging site (2 levels). However, the thing is that the imaging site factor
> only applies to only one level of the diagnosis factor (i.e. patients from
> the first level of the diagnosis factor were seen in 2 different imaging
> sites, whereas for the two other groups of patients the same imaging site
> was used).
>
> I have followed the tutorials but I still be am confused concerning how I
> should go about creating the design matrix with 3 fixed factors. Would it
> be:
>
> GroupDescriptorFile 1
> Title xx
> Class Diagnosis1-Male-Site1
> Class Diagnosis1-Female-Site1
> Class Diagnosis1-Male-Site2
> Class Diagnosis1-Female-Site2
> Class Diagnosis2-Male-Site1
> Class Diagnosis2-Female-Site1
> Class Diagnosis3-Male-Site1
> Class Diagnosis3-Female-Site1
> Inputetc etc...
>
> Afterwards, if I'm looking for a difference between Diagnosis1 and
> Diagnosis2, what would be the contrast?
> 0.25 0.25 0.25 0.25 -0.5 0.5 0 0 ?
>
> Thank you very much for your help,
>
> Shady
>
> --
> *Shady Rahayel*
> Étudiant au Ph.D. en neuropsychologie R/I
> Université du Québec à Montréal
>
> Centre d'Études Avancées en Médecine du Sommeil
> Hôpital du Sacré-Coeur de Montréal
>
> ___
> 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 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] Cluster Forming Threshold

[Anderson M. Winkler]

Hi Tim,

There is no established criterion and a recent, useful guideline is perhaps
use the threshold that would correspond to a p=0.001, that is, about 3.1 in
a z-map (see Woo et al., Neuroimage, 2014
).

All the best,

Anderson


On 1 September 2016 at 18:57, Timothy Hendrickson  wrote:

> Freesurfer Experts,
>
> I am currently using a Monte Carlo Z-Simulation to perform multiple
> comparisons on a data set.
> I have played around with various cluster forming threshold and cluster
> wise thresholds.
> I have no idea what would be considered "too liberal" or "too
> conservative" of thresholds and I have not found much literature that
> decimates this either.
> Are there any general best practices I should follow when determining
> these thresholds?
>
> Respectfully,
>
> -Tim
>
> Timothy Hendrickson
> Department of Psychiatry
> University of Minnesota
> Mobile: 507-259-3434 (texts okay)
>
> ___
> 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 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] design matrix palm

[Anderson M. Winkler]

Hi Maaike,

I know palm! :)

Please see below:

On 7 September 2016 at 09:34, maaike rive  wrote:

> Hi all,
>
> I was adviced to use PALM for permutation of non-orthogonal surface data.
> However, I have some questions  regarding the design matrix en contrast
> files to be used in PALM, but somehow I cannot subscribe to the FSL mailing
> list, and I was hoping one of the freesurfer users could help me out.
>
> I have a complex design and I would like to test for interactions as well
> for subgroups; in addition, I have several covariates. If I use the
> following collumns:
>
> A1 B1 A2 B2 cov1 cov2 cov3
> I assumed I can at least test for the effects of A vs B (1 - 1 1 -1 0 0 0)
> and 1 vs 2 (1 1 -1 -1 0 0 0) and interactions (1 -1 -1 1 0 0 0).
>
> Q1:
> Is it true that I cannot compare subgroups, because columns denoted zero
> will be handeled as being a covariate? i.e. the contrast 1 -1 0 0 0 0 0 is
> inappropriate and I would have to build a new design matrix (A1 B1 cov1
> cov2 cov3)? This is what I was told previously when using randomise for DTI
> data, but I'm not sure PALM works the same way.
>

I wouldn't say it's inappropriate, but the two models will lead to
different results. Arguably the larger model will have better estimates for
the coefficients for the three nuisance variables, and if the variances are
the same for all subjects (as assumed anyway), the variance estimates are
also better. This means that in the larger model you would use the
information from the other subjects that are not part of the contrast.


>
> Q2:
> What I also understood from randomise is that you have to define both 1 -1
> and -1 1; there is no option for a 2tailed t-test or F test. Is that
> correct and does it also apply to PALM?
>

In randomise F-tests can be used to test the two tails. In PALM, the
F-tests only work for contrasts that have rank > 1, otherwise a two-tailed
test can be done with the option -twotail. Another possibility in PALM is
to, instead of using F-tests or -twotail, use the option -corrcon, that
will correct across all contrasts.

All the best,

Anderson



>
> Thanks,
>
> Maaike
>
>
> ___
> 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 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] Display FSL PALM(permutation) stats in freeview

[Anderson M. Winkler]

Hi Ajay,

To open the .mgz files produced by PALM you would first load the surface in
FreeView (at the top of the left panel, click on "Surface" then on the
button with a "+" sign), then after the surface has been loaded, in the
menu "Overlay" in the left panel, select "Load generic..." and load the
.mgz file you'd like to see. After loading there will be a button
"Configure overlay" that opens a window where the colourbar can be set,
etc. For 0.05 the threshold would be -log10(0.05) = 1.301.

Hope this helps.

All the best,

Anderson


On 15 September 2016 at 19:46, Ajay Kurani  wrote:

> Hello Freesurfer Experts,
>I was running permutation simulations on cortical thickness data and I
> had an issue with non-orthogonal covariates with mri_glmfit-sim -perm.  I
> then tried FSL's PALM which is an extension of randomize to calculate
> threshold free stats.  I saved the output as logp(which is similar to qdec
> I believe), however I have not been able to load the stats files
> correctly.  The output of palm is lh.thickness_tfce.mgz for my various
> contrasts.
>
> 1) Is .mgz the proper format for the stats files or do I need to convert
> this to another type like .mgh etc?
>
> 2) Can I display this in freeview or is another program needed?  I also
> tried tksurfer but when I loaded the stats file as an overlay nothing
> displayed.  I want to make sure that the stats is loaded as an overlay in
> freeview/tksurfer and if so, do I need to select anything special so that
> it scales the logp values correctly?
>
> Thanks,
> Ajay
>
> ___
> 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 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] area and smoothing

[Anderson M. Winkler]

Hi Seung-Gul,

Please see below:

On 16 September 2016 at 00:52, Seung Gul Kang  wrote:

> Dear Freesurfer experts,
>
> 1. I want to compare the area and/or areapial between two groups in qdec
> menu.
> I think that the eTIV or total surface area should be used as a covariate.
> Which is more appropriate between eTIV and total surface area?
>

Using and not using a global measurement tell each different things, and a
suggestion (I'd say a recommendation) is to run both ways, i.e., with and
without global area as nuisance, and use the results from both when reading
and interpreting the findings.

Regarding ICV, it has substantial correlation with area and it can be
included as nuisance in the same model in which global area is included. It
is not a problem that they are correlated as both go as nuisance in the
group comparison.



>
> 2. Is there any guideline or rule for the smoothing in the thickness,
> area, and volume comparison?
> How about the smoothing=5?
>

There is no specific rule. We have investigated with as little as 10 mm on
the sphere and the results of the default area processing in FS become
virtually indistinguishable from the exact method that we proposed in 2012.
Although we did not investigate 5mm, we did look also into the ic5 and ic3
resolutions and, particularly for the ic5 (that has vertices about 4 times
farther from each other compared to the default ic7), 10 mm smoothing is
fine, suggesting that 5 mm would likewise lead to similar results for the
ic7. A summary of these findings is in a manuscript that we have just
uploaded to bioRxiv: http://dx.doi.org/10.1101/074666 (for this particular
question the SI is probably more important).

All the best,

Anderson




>
> Thank you.
>
> Best,
> Seung-Gul
>
> ---
> *Seung Gul Kang, M.D., Ph.D. *
>
> *Psychiatrist, Associate Professor*; Department of Psychiatry, Gil
> Medical Center, Gachon University, School of Medicine, 21, Namdong-daero
> 774 beon-gil, Namdong-gu, Incheon, 21565, South Korea
> *Research scholar*; Sleep Disorders Clinical Research Program, Department
> of Psychiatry, Massachusetts General Hospital, Harvard Medical School, 1
> Bowdoin Square, 9th floor, Boston, MA 02114, USA
> *Collaboration researcher*; Division of Sleep & Circadian Disorders,
> Brigham & Women's Hospital, Harvard Medical School, 221 Longwood Ave,
> Boston MA 02115
> ᐧ
>
> ___
> 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 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] Display FSL PALM(permutation) stats in freeview

[Anderson M. Winkler]

Hi Ajay,

The mask is a file of the "curvature" type, that is, it contains vertexwise
data, and should mask out the "unknown" region (that region in the medial
aspect that is not cortex and is there to ensure the topology is the same
as that of a sphere). The mask can alternatively be a .mgz file as a
pseudo-volume, still with the same number of vertices as the input data and
surface.

Although I'm not seeing your input files, I believe your mask is fine,
otherwise PALM would have failed even before starting the permutations. I'd
think there is something else going on. Could you show the full design,
contrasts, and command line used?

All the best,

Anderson


On 18 September 2016 at 04:03, Ajay Kurani  wrote:

> Hi Anderson,
>Thanks for the help.  When viewing my results they looked very
> strange.  Upon further investigation it looks as though the mask I supplied
> to PALM was a white matter mask (mask.mgh from running qdec initially)
> created when I ran qdec.  I assumed this would be the whole cortex but I
> was wrong.  Therefore it seems to only run permutation testing on the
> surface of the white matter.  Due to the fact that it is unsmoothed white
> matter, I think this is why we see some speckling bleeding through near the
> boundaries
>
> In order to do permutation testing accurately for surface based cortical
> thickness, would the mask need to be a volume file which is between the
> pial and white matter surfaces or would it just need to be the pial surface
> (lh.pial / rh.pial), or something else?  Any suggestions on the best way to
> create this?
>
> Thanks,
> Ajay
>
> On Sat, Sep 17, 2016 at 1:03 PM, Ajay Kurani 
> wrote:
>
>> Hi Anderson,
>>Thanks for the help.  When viewing my results they looked very
>> strange.  Upon further investigation it looks as though the mask I supplied
>> to PALM was a white matter mask (mask.mgh from running qdec initially)
>> created when I ran qdec.  I assumed this would be the whole cortex but I
>> was wrong.  Therefore it seems to only run permutation testing on the
>> surface of the white matter as seen in the attached photo.  Due to the fact
>> that it is unsmoothed white matter, I think this is why we see some
>> speckling bleeding through near the boundaries
>>
>> In order to do permutation testing accurately for surface based cortical
>> thickness, would the mask need to be a volume file which is between the
>> pial and white matter surfaces or would it just need to be the pial surface
>> (lh.pial / rh.pial), or something else?  Any suggestions on the best way to
>> create this?
>>
>> Thanks,
>> Ajay
>>
>>
>>
>>
>> On Thu, Sep 15, 2016 at 1:46 PM, Ajay Kurani 
>> wrote:
>>
>>> Hello Freesurfer Experts,
>>>I was running permutation simulations on cortical thickness data and
>>> I had an issue with non-orthogonal covariates with mri_glmfit-sim -perm.  I
>>> then tried FSL's PALM which is an extension of randomize to calculate
>>> threshold free stats.  I saved the output as logp(which is similar to qdec
>>> I believe), however I have not been able to load the stats files
>>> correctly.  The output of palm is lh.thickness_tfce.mgz for my various
>>> contrasts.
>>>
>>> 1) Is .mgz the proper format for the stats files or do I need to convert
>>> this to another type like .mgh etc?
>>>
>>> 2) Can I display this in freeview or is another program needed?  I also
>>> tried tksurfer but when I loaded the stats file as an overlay nothing
>>> displayed.  I want to make sure that the stats is loaded as an overlay in
>>> freeview/tksurfer and if so, do I need to select anything special so that
>>> it scales the logp values correctly?
>>>
>>> Thanks,
>>> Ajay
>>>
>>
>>
>
> ___
> 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 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] Display FSL PALM(permutation) stats in freeview

[Anderson M. Winkler]

Hi Ajay,

Make sure you have to run palm_hemisplit on the outputs first, i.e.:

palm_hemisplit bh.thickness_*

Note that the "-o bh.thickness" was a somewhat poor choice of outputs as
now it's necessary to use the underscore ("_") to avoid overwriting
original inputs; for future runs, consider something as "-o bh.results" or
similar.

Then load these split files on top of the surfaces.

The different results shown using white or pial make no sense I'm afraid.
The surfaces don't contain data, just the geometry on top of which the
results are displayed. Using a different surface on Freeview doesn't change
the results.

All the best,

Anderson


On 19 September 2016 at 17:38, Ajay Kurani  wrote:

> Hi Anderson,
>My full design contrasts are below:
>
> /ContrastName1 HC > Grp1
> /ContrastName2 HC < Grp1
> /ContrastName3 HC > Grp2
> /ContrastName4 HC < Grp2
> /ContrastName5 Grp1 > Grp2
> /ContrastName6 Grp1 < Grp2
> /ContrastName7 M > F
> /ContrastName8 M < F
> /ContrastName9 HC/Grp1 M/F Interaction
> /ContrastName10 HC/Grp2 M/F Interaction
> /ContrastName11 Grp1/Grp2 M/F Interaction
> /NumWaves 9
> /NumPoints 11
> /Matrix
> 1 1 -1 -1 0 0 0 0 0
> -1 -1 1 1 0 0 0 0 0
> 1 1 0 0 -1 -1 0 0 0
> -1 -1 0 0 1 1 0 0 0
> 0 0 1 1 -1 -1 0 0 0
> 0 0 -1 -1 1 1 0 0 0
> 1 -1 1 -1 1 -1 0 0 0
> -1 1 -1 1 -1 1 0 0 0
> 1 -1 -1 1 0 0 0 0 0
> 1 -1 0 0 -1 1 0 0 0
> 0 0 1 -1 -1 1 0 0 0
>
> My colums correspond to the following:
> EV1:HC-M
> EV2:HC-F
> EV3:Grp1-M
> EV4:Grp1-F
> EV5:Grp2-M
> EV6:Grp2-F
> EV7:Age
> EV8:Education
> EV9:Disease Severity
>
> In the folder I was running the analysis I put the lh.thickness.10mm.mgz,
> rh.thickness.10mm.mgz, lh.white (fsaverage), rh.white (fsaverage),
> lh.mask.mgh (taken from running qdec initially), rh.mask.mgh (taken from
> running qdec initially).
>
> I initially ran palm_hemimerge lh* within matlab
> Then from a terminal I ran the following command:
>
> palm -i bh.thickness.10mm.mgz -d design.mat -t design.con -o bh.thickness
> -n 500 -approx tail -corrcon -s bh.white -T -tfce2D -logp -m bh.mask.mgz
> -nouncorrected
>
> The command ran fully.  When I loaded contrast 6 I found no results on the
> pial surface, however the white matter surface (where the mask was) was
> speckled all over with no cluster.
>
> If there is a location,I can upload the stats file if that is easier.
>
> Thanks,
> Ajay
>
> On Sat, Sep 17, 2016 at 10:03 PM, Ajay Kurani 
> wrote:
>
>> Hi Anderson,
>>Thanks for the help.  When viewing my results they looked very
>> strange.  Upon further investigation it looks as though the mask I supplied
>> to PALM was a white matter mask (mask.mgh from running qdec initially)
>> created when I ran qdec.  I assumed this would be the whole cortex but I
>> was wrong.  Therefore it seems to only run permutation testing on the
>> surface of the white matter.  Due to the fact that it is unsmoothed white
>> matter, I think this is why we see some speckling bleeding through near the
>> boundaries
>>
>> In order to do permutation testing accurately for surface based cortical
>> thickness, would the mask need to be a volume file which is between the
>> pial and white matter surfaces or would it just need to be the pial surface
>> (lh.pial / rh.pial), or something else?  Any suggestions on the best way to
>> create this?
>>
>> Thanks,
>> Ajay
>>
>> On Sat, Sep 17, 2016 at 1:03 PM, Ajay Kurani 
>> wrote:
>>
>>> Hi Anderson,
>>>Thanks for the help.  When viewing my results they looked very
>>> strange.  Upon further investigation it looks as though the mask I supplied
>>> to PALM was a white matter mask (mask.mgh from running qdec initially)
>>> created when I ran qdec.  I assumed this would be the whole cortex but I
>>> was wrong.  Therefore it seems to only run permutation testing on the
>>> surface of the white matter as seen in the attached photo.  Due to the fact
>>> that it is unsmoothed white matter, I think this is why we see some
>>> speckling bleeding through near the boundaries
>>>
>>> In order to do permutation testing accurately for surface based cortical
>>> thickness, would the mask need to be a volume file which is between the
>>> pial and white matter surfaces or would it just need to be the pial surface
>>> (lh.pial / rh.pial), or something else?  Any suggestions on the best way to
>>> create this?
>>>
>>> Thanks,
>>> Ajay
>>>
>>>
>>>
>>>
>>> On Thu, Sep 15, 2016 at 1:46 PM, Ajay Kurani 
>>> wrote:
>>>
 Hello Freesurfer Experts,
I was running permutation simulations on cortical thickness data and
 I had an issue with non-orthogonal covariates with mri_glmfit-sim -perm.  I
 then tried FSL's PALM which is an extension of randomize to calculate
 threshold free stats.  I saved the output as logp(which is similar to qdec
 I believe), however I have not been able to load the stats files
 correctly.  The output of palm is lh.thickness_tfce.mgz for my various
 contrasts.

 1) Is .mgz the proper format for the stats files or do I

[Freesurfer] Download FS not working?

[Anderson M. Winkler]

Hi guys,

I've been trying in the last days to download FS but it doesn't seem to 
work:



winkler@volans:~ $ wget 
ftp://surfer.nmr.mgh.harvard.edu/pub/dist/freesurfer/5.3.0/freesurfer-Linux-centos6_x86_64-stable-pub-v5.3.0.tar.gz
--2014-01-09 20:47:12-- 
ftp://surfer.nmr.mgh.harvard.edu/pub/dist/freesurfer/5.3.0/freesurfer-Linux-centos6_x86_64-stable-pub-v5.3.0.tar.gz

   => 'freesurfer-Linux-centos6_x86_64-stable-pub-v5.3.0.tar.gz'
Resolving surfer.nmr.mgh.harvard.edu (surfer.nmr.mgh.harvard.edu)... 
132.183.202.158
Connecting to surfer.nmr.mgh.harvard.edu 
(surfer.nmr.mgh.harvard.edu)|132.183.202.158|:21... connected.

Logging in as anonymous ...


It never goes beyond this point. It isn't working for any of the FS 
versions available, neither trying from different computers or different 
networks. Neither using different download methods. Looks like an issue 
with the FTP server, not sure...


Thanks and sorry for bothering.

All the best,

Anderson

___
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] Multiple comparisons to confirm atrophy

[Anderson M. Winkler]

Hi Clare,

This is something that PALM 
can do. It seems we're in the same Dept. Please send me an email off-list
and we can try to have this sorted out.

All the best,

Anderson


On 19 November 2015 at 08:37, Clare Loane  wrote:

> Dear All,
>
> I have completed recon-all on my dataset of controls and patients. My
> patients have hippocampal/MTL atrophy and I would like to conduct multiple
> comparisons on the data to ensure that the atrophy is indeed focal in each
> patient and if not, identify where else atrophy is occurring in these
> patients. My hypothesis is that I do not expect atrophy elsewhere due to
> the disease type of these patients, but as this has never been
> confirmed outside of visual inspection before I wish to check this more
> robustly.
>
> I am not at all familiar with this kind of analysis and would like to ask
> some advice on how I should approach this? My initial instinct is to go
> with permutation testing. I would like to consider the whole brain 
> (subcortical
> and surface).
>
> Are there other issues I should be taking into account?
>
> Any advice and specific information regarding the subsequent analysis in
> freesurfer would be greatly appreciated.
>
> Many thanks in advance.
>
> Clare
>
> ___
> 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 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] Usage of --area option in estimating the vertex-wise maps of surface area

[Anderson M. Winkler]

Hi Xi-Nian,

As Doug said, they don't do the same thing. The method in the paper uses a
different kind of interpolation to allows the preservation of the amount of
area, whereas mris_preproc achieves the same globally using a Jacobian
correction. The results aren't expected to be identical at the vertex level
or for small regions.

An implementation of the method we proposed in the paper is available for
Octave and/or Matlab at http://brainder.org/download/areal
It works fine, but beware that it's quite slow.

All the best,

Anderson



2013/5/13 Douglas Greve 

>
> Hi Xi-Nian, it does not do the same thing, but it does something similar.
> Anderson found that it gave similar results in a lot of cases, but he has
> found some differences. If he's still following the FS list, maybe he can
> comment.
>
> doug
>
>
>
>
>
> On 5/11/13 9:53 PM, Xinian Zuo wrote:
>
> Hello Doug,
>
>  I am doing some association studies between functional measures and
> surface area. In using  mris_preproc, there is an option --area. Does this
> do sth similar to that as Winkler et al. (2012)?
>
>  Winkler et al., 2012. Measureing and comparing brain cortical
> surfacearea and other areal quantities.
>
>  Thank you very much.
> --
> Xi-Nian Zuo, Ph.D of Applied Mathematics (http://lfcd.psych.ac.cn)
> Google Scholar Citations:
> http://scholar.google.com/citations?user=a3-gVGMJ&hl=en
>
>
>
>
> ___
> Freesurfer mailing 
> listfreesur...@nmr.mgh.harvard.eduhttps://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
>
>
>
> ___
> Freesurfer mailing list
> Freesurfer@nmr.mgh.harvard.edu
> https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
>
>
> The information in this e-mail is intended only for the person to whom it
> is
> addressed. If you believe this e-mail was sent to you in error and the
> e-mail
> contains patient information, please contact the Partners Compliance
> HelpLine at
> http://www.partners.org/complianceline . If the e-mail was sent to you in
> error
> but does not contain patient information, please contact the sender and
> properly
> dispose of the e-mail.
>
>
___
Freesurfer 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] problem with surface area regression

[Anderson M. Winkler]

Hi Wen,

Brain volume has a good relationship with the overall surface area measured
in native space (so, irrespective to any kind of interpolation). We found
an R^2 of 0.856, which is higher even than the correlation of brain volume
with gray matter volume as measured via VBM-like methods.

However, even if the relationship between these measurements were poor,
include brain volume as a covariate hardly hurts the model, unless you have
only few degrees of freedom (say, a tiny number of subjects).

Another thing, and perhaps better, is to consider using global surface area
instead of brain volume as a covariate for a study of surface area, to
remove global and focus only on local effects.

All the best,

Anderson



2013/6/24 wen.zhang55 

> Dear Freesurfer,
>
> I have checked the mail list, none could answer my problem.
> I added up the value of all vertxes in the .area or .area.pial files, find
> that brain size did not contribute much to that vertex-sum. However, some
> experiment used total brain volume as a regreesion coefficient to remove
> brain size effect. In these case, with the comparison of surface area in
> vertex-to-vertex, is it proper to use the above vertex-sum as the
> regression coefficient to remove different brain size?
> Thank you very much!
>
> Cowen
>
> 2013-06-24
>
>
>
> wen.zhang55
>
> ___
> 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 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] winkler surf data with qdec ?

[Anderson M. Winkler]

Hi Greg,

It's ok to smooth using FS tools, as long as the different sizes of the
faces of the ico7 are taken into account. This means that, after running
"rpncalc", the "smoothdpx" command can be skipped, and the data can be
converted to vertexwise and imported into qdec, which will then smooth it
just once.

The reason for having a separate command for smoothing is just to be able
to work with facewise data directly.

Hope this helps!

All the best!

Anderson


2012/12/6 Gregory Kirk 

> Hi ,
>
> I have winkler variety surface area data that is already smoothed and
> converted to vertex wise data on my average subject. Each subject has
> a file rh.ico7.fwhm10.mgh, lh... and Doug gave me a way with mri_concat
> to assemble the data into a single file so i can run command line
> mri_glmfit
> type analysis and that worked out fine.
>
> I would like if i can to use qdec so our non command line unix oriented
> postdoc could easily run a lot of models of interest.
>
> the LGI page says you can do
>
> recon-all -s my_subject_id -qcache -measure pial_lgi
>
> and then add a file .Qdecrc
> with MEASURE1 = pial_lgi
>
> but recon-all -qcache will do smoothing and
> smoothing has already been done by his fancy specially designed method
> for the surface interpolated data and also my data is in .mgh format
> and not curv format like thickness or lgi.
>
> is there a way i can cook this goose ?
>
> thanks
>
> greg
> ___
> 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 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] Freesurfer vertex wise area reference

[Anderson M. Winkler]

Dear Mahinda,

For the mris_preproc, I think Doug posted already explaining how it works -
it uses a jacobian correction in a way that makes it mass-conservative. You
may want to search the archives. For the ROIs, the standard method is to
add up the areas of their constituent vertices, so it should be
straightforward to describe I believe. And the statistics are based on the
GLM, which is already well entrenched in the literature, with no need for
details other than describing carefully the design you used and how you
corrected for multiple testing.

About a general discussion on area, the methods, etc, then yes, that would
be the paper that you know already.

All the best,

Anderson


2013/1/17 MAHINDA YOGARAJAH 

> Hi,
>
> Sorry to be a little more specific - I am aware of freesurfer area
> references such as the WInkler reference but basically wanted a few lines,
> and the most appropriate references, that summarises the technical approach
> used for group area comparisons in qdec (using updated mris_preproc), and
> simple ROI  group area comparisons.
>
> Thanks,
>
> Mahinda
> ___
> 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 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] Antw: Re: thickness and area in aparc.stats and mri_glmfit

[Anderson M. Winkler]

Dear Wolff,

Compared to vertexwise, analysis of regions (ROIs) have different features.
If the area of true signal spans the whole region, or large parts of it,
then an ROI analysis tend to be more powerful as the noise is diluted when
the vertices that comprise the ROI are summed or averaged. However, ROI
analyses can also be less powerful if the signal does not span the whole
region, or is split across multiple regions. This alone could explain your
observation.

Does this help?

All the best!


2013/2/11 Wolff Schlotz 

>  Dear Anderson (and all),
>
> Many thanks for your reassuring response, but I am still puzzled about the
> lack of association between mean area (rh_medialorbitofrontal_area) and
> area in qdec (see area.png). I thought that technically there must be
> larger areas for vertices within medialorbitofrontal?
>
> Best wishes,
> Wolff
>
> >>> "Anderson M. Winkler"  11.02.2013 14:22 >>>
>  Dear Wolff,
>
> There is nothing wrong with your results. Your finding is one more
> confirmation that thickness and area are indeed different traits, which are
> influenced differently by different genetic and/or environmental factors,
> and should not be confused one with another. They represent different
> aspects of brain morphology and its development, and can (and should) be
> analyzed and interpreted each on its own right.
>
> It is also evidence that more power can be gained by using these two
> measurements separately, rather than mixed up as in methods that only
> measure gray matter volume.
>
> Assuming you did everything else correctly, your results look perfectly
> fine to me.
>
> All the best!
>
> Anderson
>
>
>
> 2013/2/11 Wolff Schlotz 
>
>>  Dear Freesurfer experts,
>>  I tested associations between a continuous predictor and thickness and
>> area in qdec (and command line mri_glmfit, which gives the same results)
>> and found a cluster being negatively associated with thickness
>> orbitofrontal, but nothing for area orbitofrontal. After exporting mean
>> thickness and area values from aparc.stats into Stata, consistent with my
>> expectation I found a significant negative correlation with my predictor
>> for thickness. However, I also found a significant positive association
>> between birth weight and area.
>>  To check what might be wrong I tested correlations between mean
>> medialorbitofrontal thickness from rh.aparc.thickness and thickness in qdec
>> and did the same for area. As expected, large average thickness values were
>> positively associated with thickness in medialoribotfrontal and adjacent
>> areas (see attached qdec screenshot thickness.png), but there were no
>> associations using area (see attached qdec screenshot area.png). My
>> expectatioin was that there should be positive area associations similar to
>> those for thickness.
>>  Hence my question: Is this expectation correct? If yes, why this
>> discrepancy between thickness and area?
>>  Thank you.
>>  Wolff
>>
>> ___
>> 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 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 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] Surface area computation

[Anderson M. Winkler]

Dear Shantanu,

An implementation that runs in Matlab or Octave is available here:
http://brainder.org/download/areal

I was planning to write a compiled version that could be integrated more
easily in FS. However, just by the time as the paper was published, I moved
to another country and started to work on different projects. Nonetheless,
if anyone is interested in writing this in C or C++ that could run faster
than the Matlab version, let me know and I'd be happy to help!

All the best,

Anderson


2013/2/15 Shantanu Ghosh 

> Hello FreeSurfer experts,
>
> Has algorithm described in Winkler et al. (2012) Neuroimage paper for
> computation of surface area been incorporated in v5.1, or is it part of
> the 5.2b implementation?
>
> Thanks for your help.
>  Shantanu
>
>
> --
> Shantanu Ghosh, Ph.D.
> Harvard Medical School & Massachusetts General Hospital
> Martinos Center for Biomedical Imaging
>
> ___
> 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 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] download freesurfer

[Anderson M. Winkler]

Hi Merlin,

Have you tried "wget -c" to resume an interrupted download? You can put the
command inside an infinite loop that will keep retrying even if the
connection is interrupted when you are not watching (say, overnight). The
following should work:

while (true) ; do wget -c -T 60
ftp://surfer.nmr.mgh.harvard.edu/pub/dist/freesurfer/5.1.0/freesurfer-Linux-centos4-stable-pub-v5.1.0.tar.gz;
sleep 600 ; done

If it doesn't receive data for 60 seconds, it stops, sleeps for 10 min, and
tries automatically again. When the download is complete, it will continue
trying again every 10 min until you press Ctrl+C, but since the file will
already be complete, it will do nothing, so it's fine.

Hope this helps!

All the best,

Anderson



2013/2/15 

> Hi,
> Could somebody say to me if exists some form  by means of which I can
> download the file "freesurfer-Linux-centos4-stable-pub-v5.1.0.tar.gz" in
> several smaller parts?
> Greetings,
> Merlin Verdecia
>
>
> --
>
> Este mensaje le ha llegado mediante el servicio de correo electronico que
> ofrece Infomed para respaldar el cumplimiento de las misiones del Sistema
> Nacional de Salud. La persona que envia este correo asume el compromiso de
> usar el servicio a tales fines y cumplir con las regulaciones establecidas
>
> Infomed: http://www.sld.cu/
> ___
> 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 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] mean thickness covariate, mean area covariate, & mri_anatomical_stats for multiple subjects

[Anderson M. Winkler]

Hi Laura,


>1. Is there a paper that I could cite that recommends using mean
>cortical thickness rather than ICV?
>
>
If it helps, we used cortical thickness and area as covariate for the
respective analysis of regional thickness and area. Brain volume, which is
more closely related to ICV, correlates well with area, but not with
thickness. We computed a global thickness average by weighting the
thickness of each region by their respective areas. The paper is this:
http://surfer.nmr.mgh.harvard.edu/ftp/articles/Winkler2010_Neuroimage.pdf


>
>1. Would the same logic be applied to surface area analyses? i.e.
>would it make more sense to use mean surface area as a covariate in surface
>area analyses? If so, which mean surface area calculation should be used?
>mri_anatomical_stats can produce both pial and white matter mean surface
>area stats.
>
>
Yes, I think so. It seems more logical to have a global measurement of area
in the model than a measurement of brain volume. On the other hand, area
and thickness are not correlated one to another (as shown in the paper
above and also in Panizzon et al, 2009, in Cereb Cortex). I don't think
there is a clear answer on which, pial or white, should be used. I'd
probably go with the white, as I think it may be more robust to image
quality, but I admit this is a rather weak justification and if the images
are good, perhaps the pial could be just as good, despite the fact that it
somewhat depends on the white for its construction.



>
>1. Is there a way to run mri_anatomical_stats on multiple subjects at
>once and write to a tablefile (similar to asegstats2table output)?
>
>
I think you can use aparcstats2table, then add up all regions in a
spreadsheet (or even with awk/gawk). Alternatively, you can use "grep" to
pick the WhiteSurfArea for each hemisphere from the ?h.aparc.stats file for
each subject.

Hope this helps!

All the best,

Anderson
___
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] mean thickness covariate, mean area covariate, & mri_anatomical_stats for multiple subjects

[Anderson M. Winkler]

Hi Laura,
I think yes, sounds right. Maybe run both ways and see if the results agree
or differ.
All the best,
Anderson


2013/3/9 Laura M. Tully 

> Hi Anderson,
>
> Thanks, that reference is particularly helpful.
>
> Re: the usage of the white versus pial area question - I believe that the
> default area calculation in freesurfer is the white surface area, so unless
> one specifies the pial in calculations, the standard surface area output
> for surface area by parcellation will be white. This suggests the use of
> the global measurement of white surface area as a covariate would be an
> appropriate, whereas if one was specifically using pial surface area in the
> aparc calculations, it may make more sense to use the global measure of
> pial surface area as a covariate, correct? As for which one to use in
> analysis, I'm not sure - conceptually it might be that the pial surface
> area is more sensitive to atrophy but I don't know if that is born out in
> the data...
>
> Laura.
>
>
> On Sat, Mar 9, 2013 at 11:49 AM, Anderson M. Winkler <
> wink...@fmrib.ox.ac.uk> wrote:
>
>> Hi Laura,
>>
>>
>>
>>>1. Is there a paper that I could cite that recommends using mean
>>>cortical thickness rather than ICV?
>>>
>>>
>> If it helps, we used cortical thickness and area as covariate for the
>> respective analysis of regional thickness and area. Brain volume, which is
>> more closely related to ICV, correlates well with area, but not with
>> thickness. We computed a global thickness average by weighting the
>> thickness of each region by their respective areas. The paper is this:
>> http://surfer.nmr.mgh.harvard.edu/ftp/articles/Winkler2010_Neuroimage.pdf
>>
>>
>>>
>>>1. Would the same logic be applied to surface area analyses? i.e.
>>>would it make more sense to use mean surface area as a covariate in 
>>> surface
>>>area analyses? If so, which mean surface area calculation should be used?
>>>mri_anatomical_stats can produce both pial and white matter mean surface
>>>area stats.
>>>
>>>
>> Yes, I think so. It seems more logical to have a global measurement of
>> area in the model than a measurement of brain volume. On the other hand,
>> area and thickness are not correlated one to another (as shown in the paper
>> above and also in Panizzon et al, 2009, in Cereb Cortex). I don't think
>> there is a clear answer on which, pial or white, should be used. I'd
>> probably go with the white, as I think it may be more robust to image
>> quality, but I admit this is a rather weak justification and if the images
>> are good, perhaps the pial could be just as good, despite the fact that it
>> somewhat depends on the white for its construction.
>>
>>
>>
>>>
>>>1. Is there a way to run mri_anatomical_stats on multiple subjects
>>>at once and write to a tablefile (similar to asegstats2table output)?
>>>
>>>
>> I think you can use aparcstats2table, then add up all regions in a
>> spreadsheet (or even with awk/gawk). Alternatively, you can use "grep" to
>> pick the WhiteSurfArea for each hemisphere from the ?h.aparc.stats file for
>> each subject.
>>
>> Hope this helps!
>>
>> All the best,
>>
>> Anderson
>>
>>
>
>
> --
> --
> Laura M. Tully, MA
> Social Neuroscience & Psychopathology, Harvard University
> Center for the Assessment and Prevention of Prodromal States, UCLA Semel
> Institute of Neuroscience
> ltu...@mednet.ucla.edu
> ltu...@fas.harvard.edu
> 310-267-0170
> --
> My musings as a young clinical scientist:
> http://theclinicalbrain.blogspot.com/
> Follow me on Twitter: @tully_laura
>
___
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] Multiple comparison question

[Anderson M. Winkler]

Hi VM,

Please see below:


On 1 February 2017 at 02:38, neuroimage analyst <
neuroimage.anal...@gmail.com> wrote:

> Hi,
>
> We extracted volume measures from 7 ROIs in 2 groups and compared
> a) Mean of each ROI volume between group, and compared independently such
> as
> i) whether mean of ROI 1 in group 1 is equal to mean of ROI1 in group 2
> ii)whether mean of ROI 2 in group 1 is equal to mean of ROI2 in group 2
>  and so on
>

This means 7 comparisons, or 14 if you test both directions (tails)
separately.


>
> b) Plotted the mean ROI volume for 2 independent measures and compared
> whether the slope between and within group is significantly different. such
> as
> i) Slope of ROI1 is positively associated with MoCA , age (grp1) ?
> ii) Slope of ROI1 is negatively associated with MoCA, age (grp2) ?
> iii) Slope of ROI1 vs MoCA (grp1) is significantly different than slope of
> ROI1 vs MoCA (grp2) ?
> iv) Slope of ROI1 vs MoCA (grp2) is significantly different than slope of
> ROI2 vs MoCA (grp2) ?
>

These sound tests for main effects and interaction, with various
combinations possible.


>
> Q1) Do we have to perform multiple comparisons for (a) and (b)?
>

Yes.



> Q2) If yes, do we correct across 7 measurements for (a) and 2 measurements
> for (b) or 7 measurements for both (a) and (b)?
>

You could put the values for the 7 ROIs in a table in .csv format, with one
column per ROI and one row per subject. This would be the input data.
Create a design matrix as:

EV1: 0 or 1 coding for group 1
EV2: 0 or 1 coding for group 2
EV3: MoCA group 1
EV4: MoCA group 2
EV5: Age group 1
EV6: Age group 2
etc

Then define a set of contrasts such as:

C1: [1 -1 0 0 0 0 ...] - This tests if mean for group 1 > mean for group 2
C2: [-1 1 0 0 0 0 ...] - This tests if mean for group 1 < mean for group 2
C3: [0 0 1 -1 0 0 ...] - This tests if the slope for MoCA for group 1 >
slope for MoCA for group 2
C4: [0 0 -1 1 0 0 ...] - This tests if the slope for MoCA for group 1 <
slope for MoCA for group 2
C5: [0 0 1 0 0 0 ...] - This tests if slope for MoCA for group 1 > 0
C6: [0 0 -1 0 0 0 ...] - This tests if slope for MoCA for group 1 < 0
etc

You'd run this analysis in PALM with something as:

*palm -i input.csv -d design.mat -t design.con -corrcon -o myresults -logp*


Hope this helps!

All the best,

Anderson



> There is evidence to both performing and nonperforming multiple
> comparisons In the literature.
>
> Any response (hopefully pointing to a literature) will be greatly
> appreciated.
>
> Thanks
>
> Regards
>
> --VM
>
> ___
> 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 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] Multiple comparison question

[Anderson M. Winkler]

Hi VM,

They can stay all in the same .csv file. The "fwep" files will have
p-values corrected across all columns of this input.

All the best,

Anderson


On 2 February 2017 at 12:09, neuroimage analyst <
neuroimage.anal...@gmail.com> wrote:

> Thanks, Anderson. If there are several other measures such as volume,
> thickness, FA,  can I then put them in a single csv file and run the
> command or I have to create 3 csv files, one each for volume, thickness,and
> FA and also invoke -corrmod?
>
> Thanks
>
> Regards
> VM
>
> On Feb 2, 2017 1:01 AM, "Anderson M. Winkler" 
> wrote:
>
> Hi VM,
>
> Please see below:
>
>
> On 1 February 2017 at 02:38, neuroimage analyst <
> neuroimage.anal...@gmail.com> wrote:
>
>> Hi,
>>
>> We extracted volume measures from 7 ROIs in 2 groups and compared
>> a) Mean of each ROI volume between group, and compared independently such
>> as
>> i) whether mean of ROI 1 in group 1 is equal to mean of ROI1 in group 2
>> ii)whether mean of ROI 2 in group 1 is equal to mean of ROI2 in group 2
>>  and so on
>>
>
> This means 7 comparisons, or 14 if you test both directions (tails)
> separately.
>
>
>>
>> b) Plotted the mean ROI volume for 2 independent measures and compared
>> whether the slope between and within group is significantly different. such
>> as
>> i) Slope of ROI1 is positively associated with MoCA , age (grp1) ?
>> ii) Slope of ROI1 is negatively associated with MoCA, age (grp2) ?
>> iii) Slope of ROI1 vs MoCA (grp1) is significantly different
>> than slope of ROI1 vs MoCA (grp2) ?
>> iv) Slope of ROI1 vs MoCA (grp2) is significantly different than slope of
>> ROI2 vs MoCA (grp2) ?
>>
>
> These sound tests for main effects and interaction, with various
> combinations possible.
>
>
>>
>> Q1) Do we have to perform multiple comparisons for (a) and (b)?
>>
>
> Yes.
>
>
>
>> Q2) If yes, do we correct across 7 measurements for (a) and 2
>> measurements for (b) or 7 measurements for both (a) and (b)?
>>
>
> You could put the values for the 7 ROIs in a table in .csv format, with
> one column per ROI and one row per subject. This would be the input data.
> Create a design matrix as:
>
> EV1: 0 or 1 coding for group 1
> EV2: 0 or 1 coding for group 2
> EV3: MoCA group 1
> EV4: MoCA group 2
> EV5: Age group 1
> EV6: Age group 2
> etc
>
> Then define a set of contrasts such as:
>
> C1: [1 -1 0 0 0 0 ...] - This tests if mean for group 1 > mean for group 2
> C2: [-1 1 0 0 0 0 ...] - This tests if mean for group 1 < mean for group 2
> C3: [0 0 1 -1 0 0 ...] - This tests if the slope for MoCA for group 1 >
> slope for MoCA for group 2
> C4: [0 0 -1 1 0 0 ...] - This tests if the slope for MoCA for group 1 <
> slope for MoCA for group 2
> C5: [0 0 1 0 0 0 ...] - This tests if slope for MoCA for group 1 > 0
> C6: [0 0 -1 0 0 0 ...] - This tests if slope for MoCA for group 1 < 0
> etc
>
> You'd run this analysis in PALM with something as:
>
> *palm -i input.csv -d design.mat -t design.con -corrcon -o myresults -logp*
>
>
> Hope this helps!
>
> All the best,
>
> Anderson
>
>
>
>> There is evidence to both performing and nonperforming multiple
>> comparisons In the literature.
>>
>> Any response (hopefully pointing to a literature) will be greatly
>> appreciated.
>>
>> Thanks
>>
>> Regards
>>
>> --VM
>>
>> ___
>> 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 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
> Hel

Re: [Freesurfer] Error viewing corrected results

[Anderson M. Winkler]

Hi all,

That's exactly as Antonin says -- I have very little to add :-)

Only a few suggestions:

- With surfaces, both cluster and TFCE statistics tend to be slow. Consider
using the tail approximation ("-approx tail -n 500 -nouncorrected")

- Include -logp, so that the p-values are in log-10 scale. Significant
p-values are then those above 1.3 (i.e., -log10(0.05). This will help to
make the figures nicer later.

All the best,

Anderson



On 8 March 2017 at 00:19, Antonin Skoch  wrote:

> Dear Sahil,
>
> I suppose, for qcache 1.3 the equivalent cluster-forming threshold z-value
> is
>
> two-tailed test:
> qnorm(1-10^-1.3/2)=1.958949
>
> for one-tailed test:
> qnorm(1-10^-1.3)=1.643704
>
> (qnorm is R function call for quantile function of normal distribution,
> you can compute this by using other methods or use statistical z-tables)
>
> And, the directionality of the hypothesis is I suppose specified by the
> sign of your contrast vector, as I wrote in my previous mail.
>
> As for the output files, you can look at the documentation:
>
> https://fsl.fmrib.ox.ac.uk/fsl/fslwiki/PALM/UserGuide#Output_files
>
> For example, if you are looking for the p-values, used cluster extent
> inference and used t-contrast, the file with FWER-corrected p-values would
> be something like
>
> output_basename_clustere_tstat_fwep.mgz
>
> Antonin
>
>
>
>
> Hello Martin and Antonin,
>
> I was following this conversation very closely to understand how to use
> PALM in FreeSurfer.
>
> Can any of you please confirm in case I am interested in checking
> correlation between gyrification index (LGI) and behavioral measure using
> two tailed, p < 0.05:
> Step 1: I used --cache 1.3
> Step 2: Because (1-10^-1.3)= 0.95, so I will have to use -C 0.95 in palm
> command
>
> Could you please confirm if thats correct and the output *_tstat.mgz is the
> final two-tailed corrected significant correlation map between LGI and
> behavioral data?
>
> Thanks a lot for this wonderful discussion.
> Sahil
>
> PS: For one-tailed: it will be -C -0.95 in palm command, correct?
>
>
>
> On Tue, Mar 7, 2017 at 3:48 PM, Antonin Skoch  wrote:
>
> > Dear Martin,
> >
> > after -s option, there have to be 2 arguments, as I specified in my previous
> > mail:
> >
> > -s fsaverage/surf/lh.white fsaverage/surf/lh.white.avg.area.mgh
> >
> > And beware that -C has to have negative sign, if your hypothesis is
> > one-tailed negative.
> >
> > Antonin
> >
> >
> >
> > Hi Antonin,
> >
> > Thank you so much for this detailed explanation, that's really useful.
> >
> > Following your instructions, I ran:
> >
> > palm -i lh.MEQ_LGI.10.mgh -s fsaverage/surf/lh.white.avg.area.mgh -d
> > check.csv -t Contrast_MEQ.csv -n 5000 -m lh.MEQ_LGI.glmdir/mask.mgh -o
> > myresults -Cstat extent -C 3.719016
> >
> > but I am getting following error:
> >
> > Running PALM alpha104 using MATLAB 9.0.0.341360 (R2016a) with the following
> > options:
> > -i lh.MEQ_LGI.10.mgh
> > -s fsaverage/surf/lh.white.avg.area.mgh
> > -d check.csv
> > -t Contrast_MEQ.csv
> > -n 5000
> > -m lh.MEQ_LGI.glmdir/mask.mgh
> > -o myresults
> > -Cstat extent
> > -C 3.719016
> > Loading surface 1/1: fsaverage/surf/lh.white.avg.area.mgh
> > Reading input 1/1: lh.MEQ_LGI.10.mgh
> >
> > Struct contents reference from a non-struct array object.
> >
> > Error in palm_takeargs (line 1632)
> > if any(size(plm.srf{s}.data.vtx,
> > 1) == ...
> >
> > Error in palm_core (line 33)
> > [opts,plm] = palm_takeargs(varargin{:});
> >
> > Error in palm (line 81)
> > palm_core(varargin{:});
> >
> > Could you please help me in resolving this error?
> >
> > Thanks much.
> >
> > On Tue, Mar 7, 2017 at 2:55 PM, Antonin Skoch  wrote:
> >
> > > Dear Martin,
> > >
> > > input -i input file is
> > >
> > > lh.MEQ_LGI.10.mgh file in your glmdir directory (for left hemisphere).
> > >
> > > As you could read in following messages in the referenced thread in FSL
> > > discussion forum, cluster-forming threshold need to be specified in z, not
> > > in t.
> > >
> > > Therefore, you would have to select cluster forming threshold and specify
> > > it as a z score.
> > >
> > > I think that your z-score for your original mri_glmfit-sim commandline
> > > argument
> > >
> > > --cache 4 neg
> > >
> > > will be  -qnorm(1-10^-4)=-3.719016. (I am not perfectly sure since I never
> > > tried negative one-side hypothesis testing in PALM).
> > >
> > > You could also use other statistics, such as cluster mass, or TFCE. See
> > > PALM user guide.
> > >
> > > Do not include -pmethodp none and -pmethodr none, since you would need the
> > > partitioning due your non-orthogonal design matrix.
> > >
> > > ?h.white.avg.area.mgh file (which you will find under fsaverage directory)
> > > goes as second argument after -s option.
> > >
> > > Therefore I suppose the commandline for cluster extent inference with
> > > cluster forming threshold p=0.0001, negative one-sided hypothesis, left
> > > hemisphere, will be hopefully something like
> > >
> > > palm
> > > -i y

[Freesurfer] Postdoctoral position at the NIH/NIMH

[Anderson M. Winkler]

We are seeking enthusiastic applicants for a Post-Doctoral Fellowship
position to help with the collection and analysis of large brain-imaging
datasets. The successful candidate will use state-of-the-art artificial
intelligence methods, with the aim of better understanding psychiatric
disorders in young people with mental illness, particularly anxiety and
depression. Our goal is to understand better the causes and mechanisms of
certain psychiatric disorders, improve their definition and classification,
and ensure the best treatment can be offered to psychiatric patients.

The successful candidate will develop and apply deep learning algorithms to
multi-modal imaging datasets that include MRI (functional, structural),
EEG, MEG, and associated behavioral and clinical data. The methods
developed by the successful candidate will be used to:

- Integrate these diverse sources of information.
- Inform the construction computational models in psychiatry.
- Test the validity of such models.

Candidates with a strong computational background (e.g. PhD in Engineering,
Physics, Computer Science, Mathematics, Statistics, Computational
Neuroscience, and related areas) who are interested in brain development
and psychopathology, are particularly encouraged to apply. Requirements for
this position include:

- Strong machine learning experience;
- Programming experience in Python (preferably), or in R/Matlab/Octave;
- Experience with open source machine learning libraries such as
Scikit-learn, Theano, and/or Tensorflow;
- Excellent interpersonal and written (English) communication skills.

Background experience in psychiatry or knowledge of neuroimaging software
are not required. However, the candidate will be expected to learn some of
these topics as part of their role in our research group.

The successful candidate will work jointly with the laboratories of Drs
Daniel Pine and Argyris Stringaris, and together with Dr Anderson Winkler,
Staff Scientist. Please write to Drs Pine (pi...@mail.nih.gov), Stringaris (
argyris.stringa...@nih.gov) or Winkler (anderson.wink...@nih.gov) with your
application and CV.
___
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] problems with cortical surface area maps

[Anderson M. Winkler]

Hi Lars,

This pattern is because the faces in a geodesic sphere don't have all
identical sizes, even if the original icosahedron was perfectly regular,
i.e., a Platonic polyhedron. We describe this phenomenon here
, and give a simple
formula that addresses it (see Appendix A: Geodesic spheres and areal
inequalities).

If you run the statistical analysis on the faces or vertices, this is never
a problem, because the conventional stats, as t, F, or R^2 are all pivotal,
i.e., don't depend on the scale of the original measurements. Nonetheless,
it's something that needs to be addressed if smoothing is to be applied, as
well as for analyses in which equal-area for each measurement unit is a
(possibly implicit) assumption.

Hope this helps!

All the best,

Anderson




On 31 May 2014 15:54, Lars M. Rimol  wrote:

> Hi,
>
> I have generated surface area maps for a sample of subjects (using
> [mris_preproc --fsgd *.*  --target fsaverage --hemi *h --meas area --out
> *.mgh). When I created a mean image across all subjects (163842 x 1) and
> looked at it in tksurfer, I noticed a pattern on the surface that's a
> little troubling. Please see the attached image file (patterninmeanareaimg_
> allsubjs_unsmoothed.jpg).
>
> I sampled the curv files to ico (with *mris_preproc*) for all subjects
> and looked at them individually (I have attached a few figures). I did find
> four resampled curv.mgh's that looked unusual, but removing them did not
> change the pattern seen in the attached figure. I also looked at some of
> the individual area files resampled to ico (using make_average_subject) and
> the outliers there were identical with the ones identified using the
> resampled curv-files (perhaps unsurprisingly). All these images are
> unsmoothed.
>
> So I'm at a loss here. Any suggestions on what to do next in terms of
> trouble shooting?
>
>
>
> (PS! I have attached the resampled curv and area files for a normal
> subject in lh_104_curvsjekk_lateral.jpg and lh_104_areasjekk_lateral.jpg,
> and for an unusual subject in lh_81_UNUSUAL_curvsjekk_lateral.jpg and
> lh_81_areasjekk_lateral.jpg. The three other subjects that seemed unusual
> looked like subject 81, and all the other subjects looked like 104.)
>
>
> Thank you!
>
> --
> yours,
>
> Lars M. Rimol, PhD
> St. Olavs Hospital
> Trondheim,
> Norway
>
>
> ___
> 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 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] problems with cortical surface area maps

[Anderson M. Winkler]

Hi Lars,

I guess you are working with vertexwise rather than facewise data, is this
right? You'd need to compute then the area-per-vertex of the same sphere
used as the target for the interpolation, i.e., the ico7
(surf/?h.sphere.reg in fsaverage I think). Attached is a Matlab function
that calculates this area-per-vertex (or the area-per-face )from a surface
and saves as a curvature file, in ASCII format. It reads also in ASCII, so
you need to convert first (mris_convert). To run, it's something as this:

*srf2area lh.sphere.reg.srf ico7.area.dpv dpv*

Then multiply either each subject area map (after interpolation), or the
average map, by 4*pi*r^2 (r should be 100), and divide the result by the
area-per-vertex from the sphere that you calculated above, and divide the
result again by the number of vertices in the sphere, which should be
163842. To do this, use the attached rpncalc function, that also runs in
Matlab, and does simple maths in curvature files that are all of the same
size, using RPN notation
. Inputs are also in
ASCII. To run, it should be something like this:

rpncalc lh.white.interp.dpv 125663.706144 * ico7.area.dpv / 163842 /
lh.white.interp.corrected.dpv

The output you can convert back to cuvature binary and load with tksurfer
or FreeView.

Hope this helps!

All the best,

Anderson



On 1 June 2014 16:32, Lars M. Rimol  wrote:

>
> Hi Anderson,
>
> Thank you!  How would you suggest I address this issue when I smooth these
> images before statistical analysis? I normally use the smoothing algorithm
> implemented in FreeSurfer (mri_surf2surf).
>
>
>
>
> --
> yours,
>
> Lars M. Rimol, PhD
> St. Olavs Hospital
> Trondheim,
> Norway
>
> ___
> 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.
>
>


srf2area.m
Description: Binary data


rpncalc.m
Description: Binary data
___
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] meshes/surfaces for sub-cortical structures

[Anderson M. Winkler]

Hi Tanya,
I think the attached script should do the trick. It calls various FS
functions behind the scenes, including mri_tessellate that Bruce was
explaining. The outputs will be in a subdirectory called 'ascii' inside
each subject directory.
All the best,
Anderson


On 1 June 2014 18:22, Tanya  wrote:

>
> Thanks Bruce!
>
> Can you please explain how I can view the structure resulting from calling 
> mri_tesselate on the left hippocampus in Matlab?
>
> I tried loading it with read_surf.m but the "magic number" doesn't fit - it's 
> 16777213 instead.
>
> I messed with read_surf.m a bit and forced loading as a QUAD_FILE, but the 
> vertices matrix looks like this:
>
>  -160.2400 0 -158.7600
>  0 -159.2400 0
>  -160.0800 0 -158.7600
>  0 -159.2400 0
>  -159.9600 0 -158.7600
>  0 -159.2400 0
>
>
> 
>
> I can read this into a correct Nx3 coordinate matrix, but I'm not sure what 
> to do with the faces.
>
> Can you maybe help with this? Is there an updated read_surf.m that should 
> work?
>
>
> Thanks again!
>
>
> Tanya
>
>
>
>
>
> 
>
>
>
> sure, you could use mri_tesselate or mri_mc for this. We don't typically
> generate them as the structures don't really have a surface-based
> organization so (for example) interior points aren't represented on the
> mesh
>
>
> On Fri, 30 May 2014, Tanya wrote:
>
> > Hi Freesurfers,I'm wondering whether the meshes (list of vertices+faces)
> > corresponding to structures such as the hippocampus and amigdala are
> > available as part of the Freesurfer output?
> > In the /surf folder there are only the meshes of whole brain structures
> > (cortex, pial, WM...). Is there any utility function to generate the meshes
> > for the smaller structures?
>
>
>
> On Fri, May 30, 2014 at 1:18 PM, Tanya  wrote:
>
>> Hi Freesurfers,
>> I'm wondering whether the meshes (list of vertices+faces) corresponding
>> to structures such as the hippocampus and amigdala are available as part of
>> the Freesurfer output?
>> In the /surf folder there are only the meshes of whole brain structures
>> (cortex, pial, WM...). Is there any utility function to generate the meshes
>> for the smaller structures?
>>
>> Thank you!
>>
>
>
> ___
> 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.
>
>


aseg2srf
Description: Binary data
___
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] problems with cortical surface area maps

[Anderson M. Winkler]

oops! I forgot to attach the files needed by these two functions... see
them attached now. I hope I'm not forgetting any others...


On 1 June 2014 19:21, Anderson M. Winkler  wrote:

> Hi Lars,
>
> I guess you are working with vertexwise rather than facewise data, is this
> right? You'd need to compute then the area-per-vertex of the same sphere
> used as the target for the interpolation, i.e., the ico7
> (surf/?h.sphere.reg in fsaverage I think). Attached is a Matlab function
> that calculates this area-per-vertex (or the area-per-face )from a surface
> and saves as a curvature file, in ASCII format. It reads also in ASCII, so
> you need to convert first (mris_convert). To run, it's something as this:
>
> *srf2area lh.sphere.reg.srf ico7.area.dpv dpv*
>
> Then multiply either each subject area map (after interpolation), or the
> average map, by 4*pi*r^2 (r should be 100), and divide the result by the
> area-per-vertex from the sphere that you calculated above, and divide the
> result again by the number of vertices in the sphere, which should be
> 163842. To do this, use the attached rpncalc function, that also runs in
> Matlab, and does simple maths in curvature files that are all of the same
> size, using RPN notation
> <http://en.wikipedia.org/wiki/Reverse_Polish_notation>. Inputs are also
> in ASCII. To run, it should be something like this:
>
> rpncalc lh.white.interp.dpv 125663.706144 * ico7.area.dpv / 163842 /
> lh.white.interp.corrected.dpv
>
> The output you can convert back to cuvature binary and load with tksurfer
> or FreeView.
>
>  Hope this helps!
>
> All the best,
>
> Anderson
>
>
>
> On 1 June 2014 16:32, Lars M. Rimol  wrote:
>
>>
>> Hi Anderson,
>>
>> Thank you!  How would you suggest I address this issue when I smooth
>> these images before statistical analysis? I normally use the smoothing
>> algorithm implemented in FreeSurfer (mri_surf2surf).
>>
>>
>>
>>
>> --
>> yours,
>>
>> Lars M. Rimol, PhD
>> St. Olavs Hospital
>> Trondheim,
>> Norway
>>
>> ___
>> 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.
>>
>>
>


dpxread.m
Description: Binary data


dpxwrite.m
Description: Binary data


srfread.m
Description: Binary data


srfwrite.m
Description: Binary data
___
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] cortical surface area measurement and vertex-wise analysis

[Anderson M. Winkler]

Hi Emma,

Could you clarify what would be the disadvantages that you've heard about?
I ask because we've studied and tried to clarify many aspects of the
analysis of surface area in this paper
. Maybe
if there's something not covered there we could try to comment then.

Thanks!

All the best,

Anderson


On 10 October 2014 22:19, Emma Thompson  wrote:

> Dear Freesurfers,
> I recently heard that there are several disadvantages to using a measure
> of cortical surface area in Freesurfer, specifically due to conducting a
> vertex-wise analysis. Can someone help me understand why a measure of
> surface area using this approach might be flawed? Thanks!
>
> ___
> 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 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] cortical surface area measurement and vertex-wise analysis

[Anderson M. Winkler]

Hi Donna,

In that same paper we also comment on the differences between
expansion/contraction and absolute areal measurements assessed at each face
of the surface (i.e., facewise), which can, after interpolation, be
converted to vertexwise to facilitate analysis using current tools (e.g.,
mri_glmfit or qdec), without loss of areal quantities.

All the best,

Anderson


On 11 October 2014 04:05, Donna Dierker  wrote:

> Hi Emma,
>
> I might not be the only one who is unsure what you mean by a vertex-wise
> cortical surface area measure.  Do you mean something like what is
> illustrated in figures 2 and 3 here:
>
> http://www.pnas.org/content/107/29/13135.figures-only
>
> ... which is similar, but not identical to the local gyrification index?
>
> If so, would you use this measure on surfaces before or after registration
> to a target atlas?
>
> Donna
>
>
> On Oct 10, 2014, at 4:19 PM, Emma Thompson  wrote:
>
> > Dear Freesurfers,
> > I recently heard that there are several disadvantages to using a measure
> of cortical surface area in Freesurfer, specifically due to conducting a
> vertex-wise analysis. Can someone help me understand why a measure of
> surface area using this approach might be flawed? Thanks!
>
>
>
> ___
> 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 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] cortical surface area measurement and vertex-wise analysis

[Anderson M. Winkler]

Hi Emma,

Please, see below:

On 13 October 2014 20:11, Emma Thompson  wrote:

> Hi Anderson,
> Thank you for your paper. I guess I am confused. I have analyzed my
> structural data with qdec, specifically looking at a measure of cortical
> thickness and volume. I'm considering looking at surface area as well but
> was told it wasn't a very good measure using qdec, I have no other
> information really. Could you speak to whether or not this measure is valid
> in so far as one runs it using qdec?
>


The data that is used in qdec comes from mris_preproc, which uses
vertexwise interpolation with Jacobian correction for local
stretches/shrinkages. The amount of area is preserved globally, which is
desirable, although locally it's something difficult to achieve. If without
any smoothing, the spatial correlation between mris_preproc and the method
in the paper is about 0.65, and the correlation at a given vertex, across
subjects, varies spatially. Note, however, that the smoothing -- that is
applied by default -- should increase the correlation vastly, but I don't
have the exact numbers at hand to provide yet.

For area, as well as for volume, the method shown in the paper is really
the gold standard: it's pycnophylactic and thus guarantees that the amount
of area is preserved locally everywhere in the brain during the
interpolation. It's the same method used in GIS applications for
geoprocessing. The downside is that the current implementation, using
Matlab, is too slow (for the ico7, it can well take 10-15h per hemisphere
in a single CPU).




> Also, in your opinion, would it be redundant to look at surface area in
> addition to measures of cortical thickness and cortical volume?
>


Actually, the redundant bit here is volume. Area and thickness are
uncorrelated phenotypically and genetically, and each provide different
pieces of information (this is shown here
<http://dx.doi.org/10.1093/cercor/bhp026> and here
<http://www.sciencedirect.com/science/article/pii/S1053811909013160>).
Volume is a mixture of both, in proportions that vary across the cortical
mantle. I would not consider using volume given that thickness and area are
available. Moreover, volume is also an "areal quantity" in the sense that
it requires the same methods for interpolation as surface area itself, so
all that is said above about area applies also to cortical volumes.

Hope this helps!

All the best,

Anderson





On Sun, Oct 12, 2014 at 3:00 PM, Anderson M. Winkler  wrote:

> Hi Donna,
>
> In that same paper we also comment on the differences between
> expansion/contraction and absolute areal measurements assessed at each face
> of the surface (i.e., facewise), which can, after interpolation, be
> converted to vertexwise to facilitate analysis using current tools (e.g.,
> mri_glmfit or qdec), without loss of areal quantities.
>
> All the best,
>
> Anderson
>
>
> On 11 October 2014 04:05, Donna Dierker  wrote:
>
>> Hi Emma,
>>
>> I might not be the only one who is unsure what you mean by a vertex-wise
>> cortical surface area measure.  Do you mean something like what is
>> illustrated in figures 2 and 3 here:
>>
>> http://www.pnas.org/content/107/29/13135.figures-only
>>
>> ... which is similar, but not identical to the local gyrification index?
>>
>> If so, would you use this measure on surfaces before or after
>> registration to a target atlas?
>>
>> Donna
>>
>>
>> On Oct 10, 2014, at 4:19 PM, Emma Thompson  wrote:
>>
>> > Dear Freesurfers,
>> > I recently heard that there are several disadvantages to using a
>> measure of cortical surface area in Freesurfer, specifically due to
>> conducting a vertex-wise analysis. Can someone help me understand why a
>> measure of surface area using this approach might be flawed? Thanks!
>>
>>
>>
>> ___
>> 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 mailing list
> Freesurfer@nmr.mgh.harvard.edu
> https://mail.nmr.mgh.harvard.edu/mailman/listinfo

Re: [Freesurfer] Running FSL randomise on FS cortical thickness data {Disarmed}

[Anderson M. Winkler]

External Email - Use Caution

Hi,

It's possible to analyse longitudinal designs with PALM. Some mild
assumptions must hold, and those assumptions are different than those of an
LME model. PALM doesn't assume normality, doesn't require a specified form
for the dependence structure, but requires compound symmetry. PALM allows
non-parametric correction of FWER and is not based on numerical
optimization, so it doesn't suffer from convergence failures. On the other
hand, LME allows multiple variance components and can accommodate some
rather unwieldy designs, but it does make distributional and covariance
assumptions and offers no trivial method (other than perhaps FDR) to deal
with the multiple testing problem. It's up to you really...

All the best,

Anderson


On Mon, 29 Jul 2019 at 17:00, Greve, Douglas N.,Ph.D. <
dgr...@mgh.harvard.edu> wrote:

> I'm not sure that is a way to do this as it is quite tricky managing the
> exchangeability in longitudinal data. Maybe PALM can do it. I'm cc'ing
> Anderson in case he wants to weigh in.
>
> On 7/28/19 7:24 PM, Mark Wagshul wrote:
> >
> > External Email - Use Caution
> >
> > Doug,
> >
> > Thanks.  Unfortunately, this is only for the GLM, is there an
> > equivalent version for the linear mixed model? I haven’t seen one.  I
> > guess we could figure out how to integrate lme into mri_glmsim, but
> > I’m not sure that’s a better use of our time than manually running the
> > simulations.  Your thoughts?
> >
> > Thanks for the help.
> >
> > Mark
> > ___
> > Mark Wagshul, PhD
> > Albert Einstein College of Medicine
> > Bronx, NY
> >
> > Sent from my iPhone
> >
> > On Jul 28, 2019, at 4:11 PM, Greve, Douglas N.,Ph.D.
> > mailto:dgr...@mgh.harvard.edu>> wrote:
> >
> >> FreeSurfer has its own permutation software
> >> *MailScanner has detected a possible fraud attempt from
> >> "nam02.safelinks.protection.outlook.com" claiming to be*
> >>
> https://surfer.nmr.mgh.harvard.edu/fswiki/FsTutorial/MultipleComparisonsV6.0Perm
> >> We also have a handy interface to palm
> >> *MailScanner has detected a possible fraud attempt from
> >> "nam02.safelinks.protection.outlook.com" claiming to be*
> >> https://surfer.nmr.mgh.harvard.edu/fswiki/FsPalm
> >>
> >> On 7/27/2019 9:33 PM, Mark Wagshul wrote:
> >>>
> >>> External Email - Use Caution
> >>>
> >>> Great, thanks!
> >>>
> >>> Mark
> >>> ___
> >>> Mark Wagshul, PhD
> >>> Albert Einstein College of Medicine
> >>> Bronx, NY
> >>>
> >>> Sent from my iPhone
> >>>
> >>> On Jul 27, 2019, at 2:43 PM, Antonin Skoch  >>> > wrote:
> >>>
>  External Email - Use Caution
> 
>  Dear Mark,
> 
>  I would suggest to use PALM instead:
> 
>  *MailScanner has detected a possible fraud attempt from
>  "nam02.safelinks.protection.outlook.com" claiming to be*
>  https://fsl.fmrib.ox.ac.uk/fsl/fslwiki/PALM/UserGuide  <
> https://nam02.safelinks.protection.outlook.com/?url=https%3A%2F%2Ffsl.fmrib.ox.ac.uk%2Ffsl%2Ffslwiki%2FPALM%2FUserGuide&data=02%7C01%7Cmark.wagshul%40einstein.yu.edu%7Ca32b0472033a420f1ea508d71397c21a%7C04c70eb48f2648079934e02e89266ad0%7C1%7C0%7C636999414844352500&sdata=wODbvJGuGHP8dlfhO4tJjX8Ype0ph%2FbXrjMkTsGFfD8%3D&reserved=0
> >
> 
>  PALM supports surface-based data. Each vertex has different area,
> which has to be accounted when the cluster size is computed.
> 
>  Antonin Skoch
> 
> 
> 
>  Dear Freesurfer experts,
>  Hi.  We are analyzing cortical thickness data in Freesurfer,
> longitudinal data
>  in 55 subjects using the linear mixed effects package.  We are
> finding a number
>  of siginificant clusters, and would like to correct for multiple
> comparisons
>  with permutation testing.  We have already run about 2000
> permutations, and
>  would like to use TFCE available in randomise rather than empirical
> methods to
>  set parameters such as the cluster forming threshold.
> 
>  So, the simple question is, how do we convert our FS format data
> (these are F
>  statistics at each vertex) into a format which can be read into
> randomise?
> 
>  Thanks for any advice you can provide.
> 
>  Best,
> 
>  Mark
>  ___
>  Freesurfer mailing list
>  Freesurfer@nmr.mgh.harvard.edu  >
> 
> https://nam02.safelinks.protection.outlook.com/?url=https%3A%2F%2Fmail.nmr.mgh.harvard.edu%2Fmailman%2Flistinfo%2Ffreesurfer&data=02%7C01%7Cmark.wagshul%40einstein.yu.edu%7Cb7ea74f6c51b44fac3a508d712c25393%7C04c70eb48f2648079934e02e89266ad0%7C1%7C0%7C636998498148426272&sdata=hwa50qaC%2FS3Ma1o%2Bwy20l99PJ3mInnB%2FJqcX192PWvY%3D&reserved=0
>  <
> https://nam02.safelinks.protection.outlook.com/?url=https%3A%2F%2Fmail.nmr.mgh.harvard.edu%2Fmailman%2Flistinfo%2Ffreesurfer&data=02%7C01%7Cmark.wagshul%40einstein.yu.edu%7Ca32b0472033a420

[Freesurfer] Sources Sought: Octave/MATLAB programmer {Disarmed}

[Anderson M. Winkler]

External Email - Use Caution

Dear all,

The NIMH/NIH is looking for a skilled Octave/MATLAB programmer, with
knowledge of multivariate statistics, to implement inference for canonical
correlation analysis (CCA) using permutations, as described in the recent
paper 
(https://secure-web.cisco.com/1OvbxEv9_JHQute8AUV1-CRpI494R210wv39uPXMnlrig76REwW2lxED8uyMCG7YCdz4Mg9Bnx-MAV8G21jJr3Sb9iOnSDo91FrVA3B0YU1sSPf4zBxKNPe1lQNXEbYcvuvsT_l6wPHVNuL-vncgAwFCjNCxunhWNjWxPMaKscZ9fFyry-xu7zmOqKBGGnsFPMCGLgNT_UwK-4Y7yNvXCyLpIjz2zDHCMbR0wHO__fOiVgrKsBAkNwADlNOTDcs02Wi0bQeP_2Cy2LQlEkQk-xw/https%3A%2F%2Fdoi.org%2F10.1016%2Fj.neuroimage.2020.117065),
 together with
other features that have already been published and which are part of PALM,
as described in the "Sources Sought" notice linked below:

https://secure-web.cisco.com/1fHXlDSQEyeCMPT25xy9Gg_S11v-upDxypVxM2AtRSYwhk9YLYyqlgjHUhWkVHyIdC_JPiHePTSL2HYfYqi37T-e0iJ9DbroAi_wBWkOJgEwn0YF5p1c35mHCuM1St8xC2Mn-M6Z9PYWbLcv_-LpT0krKslKgT5x5fjAy_8pbuizmAirFwxeOszGMk4aqIhbllZuMCwEBc4-Zkx_IWONkvrUvzQBewamMxgS-U4R-5ZoO5i73WwEjbrP0Uk9ONDy6FhS-NbuaHYJRHTpSxyHylA/https%3A%2F%2Fbeta.sam.gov%2Fopp%2Fcda6dab7027348cf876c8e048ceffcc6%2Fview

Note:
- The search is open to persons or businesses.
- The search is not limited to US citizens or to vendors based in the US.
- The prospective vendor must have a Data Universal Numbering System (DUNS)
number and an NATO Commercial and Government Entity Code (NCAGE code).
- The prospective vendor must have either a Social Security Number (SSN) or
a Tax Identification Number (TIN).
- This is not employment or scholarship, but provision of a specific
service to the US government.

Please, read thoroughly the Sources Sought page in the link above. If you
have availability and capability to perform this potential requirement,
please contact Mr. Thien Nguyen (thien.nguy...@nih.gov, +1-301-827-0914).

All the best,

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

[Freesurfer] Orthographic camera in FreeView

[Anderson M. Winkler]

External Email - Use Caution

Hi all,

Is there a way to configure the camera in the 3D viewport in FreeView to
use an orthographic projection instead of perspective (which currently is
the default)?

Orthographic used to be the default in tksurfer. A nice feature about it is
that when we zoom, the effect is the same as scaling the projection (as
opposed to introducing distortions). It makes a difference when we are
collating screen captures for a figure, for example.

I checked the help but don't seem to find any related option (in 7.2).

Thanks!

All the best,

Anderson
___
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 Mass General Brigham 
Compliance HelpLine at https://www.massgeneralbrigham.org/complianceline 
 .
Please note that this e-mail is not secure (encrypted).  If you do not wish to 
continue communication over unencrypted e-mail, please notify the sender of 
this message immediately.  Continuing to send or respond to e-mail after 
receiving this message means you understand and accept this risk and wish to 
continue to communicate over unencrypted e-mail. 

Re: [Freesurfer] Orthographic camera in FreeView

[Anderson M. Winkler]

External Email - Use Caution

Hi Ruopeng,

Thank you for adding this feature! Much appreciated!

It doesn't seem to have worked quite yet, though:

*freeview: symbol lookup error:
/gpfs/gsfs6/users/EDB/opt/freesurfer/7.2.0/lib/qt/plugins/platforms/../../lib/libQt5XcbQpa.so.5:
undefined symbol: FT_Get_Font_Format*

I tried in two ways: (a) replacing freeview from 7.2.0 as per the
instructions, and also (b) running from the whole dev version. Both gave
the same error message.

Since we are at it :-) may I ask another related feature? In tksurfer, if
we had a bigger brain and a smaller brain, and opened them in separate
instances, we'd still see the bigger as bigger than the smaller. That is,
the default scaling was fixed regardless of the size of the surfaces. WIth
FreeView, it appears that by default the surfaces are scaled to cover some
fixed amount of the window (say, maybe 90% or so). It's fine as is just for
browsing, but if we call in a script and take screenshots, we have the
impression that the brains all have the same size, even when they differ
(sometimes substantially).

Thanks again!

All the best,

Anderson


On Mon, Apr 25, 2022 at 10:18 AM Wang, Ruopeng 
wrote:

> Hi Anderson,
>
> I’ve added “Orthographic view” option for the 3D viewport. It can be
> activated from right click menu in the view. You can download the latest
> build of freeview here:
>
> https://secure-web.cisco.com/1Jf501D91H-op1cklH0NEXBp4CngHCC6k-BcsMwioDfsUbaU_bffOu_8X5v2xH1F1WjtgGxCQwp3712llE8H8Qz6AOEAPuPeJo03QWo3_6tpvIkbvPETjkj-KeALjtWz71PfVBDhpY5MfBWgU4IlUdFwPw_-knEzNDs2qDWARMZgCzDbDy83KdPp2Pp-bSpB0Qzbo-fUD5VNX5j5aseiXXq6NzewV0_9fWfpWhiyJ8-e9Y5Xd0Uq4pA_dy7_uJ_5C588pMrfwckNIZ1gC4lVRadF16btGwGs9hUVgDRmhmji62KsNkNffeJPyRul0m-HnNUO4X89Epb7M4agLwMtb3A/https%3A%2F%2Fsurfer.nmr.mgh.harvard.edu%2Ffswiki%2FUpdateFreeview
>
> Best,
> Ruopeng
>
> On Apr 15, 2022, at 10:23 AM, Anderson M. Winkler 
> wrote:
>
> External Email - Use Caution
>
> Hi all,
>
> Is there a way to configure the camera in the 3D viewport in FreeView to
> use an orthographic projection instead of perspective (which currently is
> the default)?
>
> Orthographic used to be the default in tksurfer. A nice feature about it
> is that when we zoom, the effect is the same as scaling the projection (as
> opposed to introducing distortions). It makes a difference when we are
> collating screen captures for a figure, for example.
>
> I checked the help but don't seem to find any related option (in 7.2).
>
> Thanks!
>
> All the best,
>
> Anderson
> ___
> Freesurfer mailing list
> Freesurfer@nmr.mgh.harvard.edu
> https://secure-web.cisco.com/1FkMR2VGAMtfqe7ht5ISnPm9S1RHXsh7zW6Kt3yV-YIR_shuiKshgqMksgtYUAWHT7kI57GbGub79KCydERGduoXtOj5cmiyYNnRKz48HfZxaLXmbEfEUBdBWQCrC2CBo373QpawdpJwTCA0GQzopMVCC3sE6NdGVlb7-eMRnJ4v6zeJR-vaBXBC_6E6CvncJDWxCCry8r1TQvJi4bFbD9cjCmDV61iYOEkDtLgaGV2yD0ACD9I5wh7LU0pzfmS21b9l74gcLMlGP2-gxHlyrV9hpyZWgfzzdwMdZUgqKqUpy1G7cXxmeiPHL3u9cMyFiA0xnQQ2CRU0m-QGq4guILQ/https%3A%2F%2Fmail.nmr.mgh.harvard.edu%2Fmailman%2Flistinfo%2Ffreesurfer
>
>
> ___
> Freesurfer mailing list
> Freesurfer@nmr.mgh.harvard.edu
> https://secure-web.cisco.com/1FkMR2VGAMtfqe7ht5ISnPm9S1RHXsh7zW6Kt3yV-YIR_shuiKshgqMksgtYUAWHT7kI57GbGub79KCydERGduoXtOj5cmiyYNnRKz48HfZxaLXmbEfEUBdBWQCrC2CBo373QpawdpJwTCA0GQzopMVCC3sE6NdGVlb7-eMRnJ4v6zeJR-vaBXBC_6E6CvncJDWxCCry8r1TQvJi4bFbD9cjCmDV61iYOEkDtLgaGV2yD0ACD9I5wh7LU0pzfmS21b9l74gcLMlGP2-gxHlyrV9hpyZWgfzzdwMdZUgqKqUpy1G7cXxmeiPHL3u9cMyFiA0xnQQ2CRU0m-QGq4guILQ/https%3A%2F%2Fmail.nmr.mgh.harvard.edu%2Fmailman%2Flistinfo%2Ffreesurfer
> 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 Mass General
> Brigham Compliance HelpLine at
> https://secure-web.cisco.com/1GgQBeUgkDJPIdR5KeEo05IqJQvVUVAexBWU-LP6ZEt1owTBOFIfjWtr9svCKMgRcs7Y5fZWDArELQlqSMxTIXBEjOHtrKr6hXOY-jF-FdOKQXj6dO0u6QUdAWyvBfoMTK_F2EZtReIkfHHDRayWm3865SDXXjKoX4-SJ98IyD6_18BUJ11RgNYgiD_r_JG7HLgGaTXpPwwKrLnGNB21HJLjliLT8NkChLzTH1mhZzHACyNwrhAozR_iTPRmWUbmQcjOK_Hps-ZvOa4HSnNS86RR2KWScDzmUXirGshKL8Sf0JYX4irMkpxLeO-f8RVfhJ6SOZsq9obOUuz2ruzbl9w/https%3A%2F%2Fwww.massgeneralbrigham.org%2Fcomplianceline
>  <
> https://secure-web.cisco.com/1GgQBeUgkDJPIdR5KeEo05IqJQvVUVAexBWU-LP6ZEt1owTBOFIfjWtr9svCKMgRcs7Y5fZWDArELQlqSMxTIXBEjOHtrKr6hXOY-jF-FdOKQXj6dO0u6QUdAWyvBfoMTK_F2EZtReIkfHHDRayWm3865SDXXjKoX4-SJ98IyD6_18BUJ11RgNYgiD_r_JG7HLgGaTXpPwwKrLnGNB21HJLjliLT8NkChLzTH1mhZzHACyNwrhAozR_iTPRmWUbmQcjOK_Hps-ZvOa4HSnNS86RR2KWScDzmUXirGshKL8Sf0JYX4irMkpxLeO-f8RVfhJ6SOZsq9obOUuz2ruzbl9w/https%3A%2F%2Fwww.massgeneralbrigham.org%2Fcomplianceline>
>  .
>