Re: [Freesurfer] Coordinates in TRACULA group analysis

Thanks so much!!

********************************************************************** 京都府立医科大学附属病院 精神科・心療内科 渡辺 杏里 ********************************************************************** Anri WATANABE, M.D. Department of Psychiatry, University Hospital, Kyoto Prefectural University of Medicine **********************************************************************

2016-08-03 23:15 GMT+09:00 Anastasia Yendiki ayendiki@nmr.mgh.harvard.edu:

That's correct!

On Wed, 3 Aug 2016, Anri WATANABE wrote:

Hi Anastasia,

...

Thank you for your kind explanation. I can be understanding how TRACULA tracts white matter pathways. At each position (in training subjects, not in new subjects of my data) probability that next goes for which direction to every labels in the aparc+aseg (not with setting a certain ROI) is computed with use of training subjects. The prior probabilities are made from training subjects (your publication in 2011) and they are based on manual labeling (the manual labeling tracts are referred to Wakana et al. 2007). Then when white matter pathways are reconstructed in my subjects, TRACULA computes anatomical priors in each subjects in my data in pre-processing and probability distributions from anatomical priors and by fitting ball-and-stick model. Is this comprehension ok?

Best Regards, Anri

---

京都府立医科大学附属病院 精神科・心療内科 渡辺 杏里

---

Anri WATANABE, M.D. Department of Psychiatry, University Hospital, Kyoto Prefectural University of Medicine

---

2016-07-31 22:49 GMT+09:00 Anastasia Yendiki < ayendiki@nmr.mgh.harvard.edu>:

  Hi Anri - Instead of hard-coding some ROIs in the white matter that

the tract is forced to go through, TRACULA uses information like "what is the probability that this tract goes lateral/anterior/etc to XXX", where XXX any of the labels in the aparc+aseg. TRACULA computes these prior probabilities from a set of training subjects, where the tracts have been labeled manually. So it knows how likely a tract is to go through a certain aparc+aseg label, or to he left, right, anterior, etc of a certain aparc+aseg label. This is computed separately at each position along the tract. It's computed from the training subjects, and then used when reconstructing the tract in the new subject that you run TRACULA on.

  Hope this helps,
  a.y

  On Sun, 31 Jul 2016, Anri WATANABE wrote:

        Thanks, AnastasiaI know that TRACULA use probabilistic

tractography but if ROIs are not set how determine the origin and the end of a certain tract? I think that the first we have to determine the origin and the end of the tract, the second it constructs possible pathway not with the deterministic way (only 1 direction / 1 voxel) but with the probabilistic way (considering which direction should be next to). Is this comprehension wrong? Thank you.

        Anri

---

        京都府立医科大学附属病院
        精神科・心療内科
        渡辺 杏里

---

        Anri WATANABE, M.D.
        Department of Psychiatry,
        University Hospital, Kyoto Prefectural University of Medicine

---

        2016-07-31 13:22 GMT+09:00 Anastasia Yendiki <

ayendiki@nmr.mgh.harvard.edu>:

              Hi Anri - TRACULA does not use deterministic ROIs. It

uses a probabilistic model of how likely each tract is to go through or next to each of the labels of the freesurfer subcortical segmentation and cortical parcellation, as a function of position along the trajectory of the tract.

              Best,
              a.y

              On Sun, 31 Jul 2016, Anri WATANABE wrote:

                    Hi Anastasia,
                    Thank you! It seems work well!!
                    I have another question. Are ROIs for automatic

tractography in TRACULA the same ROIs presented in Wakana et al. 2007?

                    Anri

---

                    京都府立医科大学附属病院
                    精神科・心療内科
                    渡辺 杏里

---

                    Anri WATANABE, M.D.
                    Department of Psychiatry,
                    University Hospital, Kyoto Prefectural University

of Medicine

---

                    2016-07-27 13:17 GMT+09:00 Anastasia Yendiki
                    <ayendiki@nmr.mgh.harvard.edu>:

                          Hi Anri - The problem is in this line:

                                       set cmd = ($cmd --ref

$cvstempdir/$cvstemp)

                          It should be changed to this:

                                       set cmd = ($cmd --ref
                    $cvstempdir/$cvstemp/mri/norm.mgz)

                          For this to take effect, you need to run

"which trac-all" and make the change in the trac-all file that the which commands shows you.

                          Hope this helps,

                          a.y

                          On Fri, 15 Jul 2016, Anri WATANABE wrote:

                                Hi, AnastasiaThis is

trac-all.local-copy from 1 subject. Thank you!

                                Anri

---

                                京都府立医科大学附属病院
                                精神科・心療内科
                                渡辺 杏里

---

                                Anri WATANABE, M.D.
                                Department of Psychiatry,
                                University Hospital, Kyoto

Prefectural University of Medicine

---

                                2016-07-13 6:16 GMT+09:00 Anastasia

Yendiki ayendiki@nmr.mgh.harvard.edu:

                                      Hi Anri - This may be a bug

that was fixed at some point. Can you send me the scripts/trac-all.local-copy from one of your subjects? Thanks!

                                      a.y

                                      On Mon, 11 Jul 2016, Anri

WATANABE wrote:

                                            Hi Anastasia,There is an

error in .log files of left corticospinal tract in cvs template and I attached one of file. In addition .log files of right corticospinal tract in cvs template doesn't exist. Thanks in advance.

                                            Anri

---

                                            京都府立医科大学附属病院
                                            精神科・心療内科
                                            渡辺 杏里

---

                                            Anri WATANABE, M.D.
                                            Department of Psychiatry,
                                            University Hospital,

Kyoto Prefectural University of Medicine

---

                                            2016-07-07 20:12

GMT+09:00 Anastasia Yendiki < ayendiki@nmr.mgh.harvard.edu>:

                                                  Hi Anri - Is there

an error in the stats/*.log files for the different tracts?

                                                  a.y

                                                  On Tue, 5 Jul 2016,

Anri WATANABE wrote:

                                                        Dear

Anastasia,

                                                        I use TRACULA

to obtain diffusion measures at each voxel in a certain pathway for group analysis, but there aren't stats/*.path.mean.txt files. I found .log files

(<tract>_PP.avg33_mni_bbr.log) which exist 1 file per 1 tract, except corticospinal tract which has 2 .log files.

Command: trac-all –stat –c

                    $TUTORIAL_DATA/diffusion_tutorial/dmrirc.example

                                                        Error

log: Loading output reference volume from

                    /Applications/freesurfer/subjects/cvs_avg35

                                                        corRead():

can't open file

/Applications/freesurfer/subjects/cvs_avg35/COR-.info

                                                        ERROR: Could

not read

                    /Applications/freesurfer/subjects/cvs_avg35


                                                        I attached

dmrirc.example (configuration file) and <subjd>/scripts/trac-all.log.

                                                        Thanks in

advance, Anri

---

                                                        京都府立医科大学附属病院
                                                        精神科・心療内科
                                                        渡辺 杏里

---

                                                        Anri

WATANABE, M.D. Department of Psychiatry, University Hospital, Kyoto Prefectural University of Medicine

---

                                                        2016-06-03

9:51 GMT+09:00 Anri WATANABE <z2aanri@koto.kpu-m.ac.jp

...

:

                                                              Hi,

Anastasia. There aren't any .log files but text files like lh.ilf_AS.avg33_mni_bbr.FA_Avg.txt. I guess text files complete all pathways and measures.

---

                                                        京都府立医科大学附属病院
                                                        精神科・心療内科
                                                        渡辺 杏里

---

                                                        Anri

WATANABE, M.D. Department of Psychiatry, University Hospital, Kyoto Prefectural University of Medicine

---

                                                        2016-06-02

3:03 GMT+09:00 Anastasia Yendiki < ayendiki@nmr.mgh.harvard.edu>:

                                                              Thanks,

Anri. So the previous steps seem to have run fine. Are there any .log files created in the stats/ folder, which is created by trac-all -stat?

                                                              On Wed,

1 Jun 2016, Anri WATANABE wrote:

Hi Anastasia, This is a

<subjid>/scripts/trac-all.log of one subject of

the group.

Thanks,

Anri

---

                    京都府立医科大学附属病院

精神科・心療内科

渡辺 杏里

---

Anri WATANABE, M.D.

Department of Psychiatry,

University Hospital, Kyoto Prefectural University of Medicine

---

2016-05-31 22:55 GMT+09:00 Anastasia Yendiki

                                            <

ayendiki@nmr.mgh.harvard.edu>:

Hi Anri -
                    Can you also
                                            send your log file

(scripts/trac-all.log)?

I'll need to see what

exactly was
                    running
                                            when the error occurred.

Thanks!

a.y


On Sat, 28
                    May 2016,
                                            Anri WATANABE wrote:


      Hello
                                            Anastasia,sorry for few

information and let me tell you

command and

      error
                    log.


                                            Command: trac-all –stat –c


                    $TUTORIAL_DATA/diffusion_tutorial/dmrirc.example


      Error
                                            log: Loading output

reference volume from

                    /Applications/freesurfer/subjects/cvs_avg35



                    corRead(): can't
                                            open file

/Applications/freesurfer/subjects/cvs_avg35/COR-.info

      ERROR:
                    Could not
                                            read


                    /Applications/freesurfer/subjects/cvs_avg35

dmrirc.example (configuration file) is attached to this

e-mail.

      Thanks
                    in
                                            advance,

      Anri

---

                                            京都府立医科大学附属病院


                    精神科・心療内科

      渡辺
                    杏里

---

      Anri
                    WATANABE,
                                            M.D.


                    Department of
                                            Psychiatry,


                    University
                                            Hospital, Kyoto

Prefectural University of Medicine

---

                    2016-05-27 22:57
                                            GMT+09:00 Anastasia

Yendiki

                                            <

ayendiki@nmr.mgh.harvard.edu>:

                    Hi Anri -
                                            I do not know what

command line you ran and

what your

                    configuration
                                            file looks like, so it is

very

                    hard for
                                            me to suggest solutions.



                    Best,


                    a.y



                    On Fri, 27
                                            May 2016, Anri WATANABE

wrote:

                          Hi
                                            Anastasia,



                                            Thank you for your answer.



                                            There

aren't stats/*.path.mean.txt files and

terminal says 'Could

      not
                    read




                    /Applications/freesurfer/subjects/cvs_avg35.' I



                                            checked

/Application/freesurfer/subjects/cvs_avg35 folder and

                    couldn't find
                                            COR-.info file.



                                            Could you tell me any

resolutions, please?

                                            Thanks,


                          Anri

---

                                            京都府立医科大学附属病院



                                            精神科・心療内科


                          渡辺
                                            杏里

---

                          Anri
                                            WATANABE, M.D.



                                            Department of Psychiatry,



                                            University Hospital,

Kyoto Prefectural University

of Medicine

---

                                            2016-05-21 6:48 GMT+09:00

Anastasia Yendiki

                                            <

ayendiki@nmr.mgh.harvard.edu>:

                                              Hi Anri - The FA values

are extracted in

the native space

      of
                    each subject,
                                            which is why



                                            those are the only



                                              coordinates that you

see. If you want to

display the

                    results of your
                                            analysis on an average



                                            path, after



                                              running trac-all -stat,

you can use the

                                            stats/*.path.mean.txt

files (see also the last part

                          of
                                            the TRACULA



                                              tutorial).




                                              Best,



                                              a.y




                                              On Wed, 18 May 2016,

Anri WATANABE wrote:

                                                    Dear experts,



                                                    I use TRACULA to

examine a measure

(FA) at each voxel

      in one
                    pathway.

pathstats.byvoxel.txt files show

coordinates in

      native
                    space and
                                            after converting



                                            those the new



                                                    files don't show

any coordinates

which are in MNI

      space.



                                                    Could you tell me

how can I know MNI

coordinate

                    values?




                                                    Thank you!





                                                    Regards,



                                                    Anri

---

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.