I forward again my email, that has not been sent to the mailing list yet.

---------- Forwarded message ----------
From: Marco Bucci <bucci.marco@gmail.com>
Date: 2016-03-22 10:05 GMT+02:00
Subject: Re: [Freesurfer] "Bump" - Fwd: lGI value is aberrantly high and misalignment fix
To: Freesurfer support list <freesurfer@nmr.mgh.harvard.edu>

Dear Marie and Freesurfer Users,

thanks for your support.

I could not imagine that freeview was visualizing wrongly the surfaces.

For this reason, my colleague and I have added a small "integrity check"

with which we compare the min and max of the vertexes coordinates of

pial and pial-outer.

If the pial min and max are closer to the center it means that pail is included

in the outer surface.

We believe this is faster than checking visually with the viewer all the surfaces,

slice by slice.

The subject surfaces of which I sent you the recon.log passed the test.

But... I ran another case and unfortunately it did not pass the check.

I attach the recon log, the screenshot of tkmedit with lh.pial going below

the outer surface. Is it important for the lgI calculation purposes?

I also attach the Matlab code for the testing (it is rough, but works)

It is possible to check with a stepsize of 1 to be absolutely sure

about any intersection, but it takes longer time.

Looking forward to hear from you soon.

Thanks in advance for your help

Yours,

Marco (and Harri)

P.S. I went back to the original line 51 of the Marie Schaer code in "make_outer_surface". No "fix" anymore.

2016-03-17 23:19 GMT+02:00 Marie Schaer <Marie.Schaer@unige.ch>:

Hi Marco,

It’s pretty strange because from the recon-all.log you sent, the first LGI process went fine (line 14584 and 19329 for the right and left hemispheres, on Dec 15). Then you don’t run anything else on the surfaces and just reprocess the exact same LGI process for the same subject (Dec 18), and you get different values from what you just had, with many values <1, plus it gets stuck.

Could that be related to your change on line 51 in make_outer_surface? When you say misalignment, do you mean in freeview or tkmedit? Because we saw some occurrences of misalignment in freeview, which indeed are related to the surface format and as long as the surfaces align well in tkmedit then you should definitely not change anything in the make_outer_surface code.

Keep me posted,

Marie

On 16 Mar 2016, at 02:51, Marco Bucci <bucci.marco@gmail.com> wrote:

Dear all,

I forward again my original post.

I also attach the recon-all.log of one of the several faulty cases.

I run freesurfer (freesurfer-Darwin-lion-stable-pub-v5.3.0)

in Mac OS X 10.9.4 and -lgI option with Matlab_R2013a

and I have fixed the SearchProjectionOnPial.m.

I also had to fix line 51 of make_outer_surface.m

"v2=[129-v(:,1) v(:,3)-129 129-v(:,2)]; % in order to cope with the different orientation"

with
"v2=[vol.vox2ras1(1,4)-v(:,1) v(:,3)+vol.vox2ras1(2,4) vol.vox2ras1(3,4)-v(:,2)];"
in order to use the image coordinate to fix surface orientation and alignment.

If I would use the original line 51, the surface produced by recon -lgi results

misaligned with the reference pial and image.

My problems remains the aberrantly high lgI value in most of my subject images.

Can you help me?

Yours,

Marco Bucci

---------- Forwarded message ----------
From: Marco Bucci <bucci.marco@gmail.com>
Date: 2015-12-22 23:49 GMT+02:00
Subject: Re: [Freesurfer] lGI value is aberrantly high
To: "freesurfer@nmr.mgh.harvard.edu" <freesurfer@nmr.mgh.harvard.edu>

Dear all,

I have a problem with the LGI module.
At the step when the software computes the LGIs for the vertices,
In one vertex the lGI results aberrantly high and the software exits
with ERRORS.

I tried the solution suggested by Marie Schaer to rereconstruct
The images with "randomness"
Recon-all -autorecon2-wm -randomness -autorecon3

I tried this fix on two of my subjects
With errors. Not all subject resulted
Errors in the computation of the gyrification.

Can you help me?

Marco Bucci
Turku Pet Center

Sent from my iPhone

--

****************************************************
Marco Bucci, PhD in
Clinical Physiology and Nuclear Medicine
Turku PET Centre
Turku University Central Hospital (TUCH)
PL 52, Kiinanmyllynkatu 4-8, 20521 Turku, Finland
tel. +358 2 3132798
fax. +358 2 2318191
mobile: +358 40 6873976
UTU e-mail: marbuc@utu.fi
****************************************************

<recon-all.log>_______________________________________________
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.

****************************************************
Marco Bucci, PhD in

Clinical Physiology and Nuclear Medicine

Turku PET Centre
Turku University Central Hospital (TUCH)
PL 52, Kiinanmyllynkatu 4-8, 20521 Turku, Finland
tel. +358 2 3132798
fax. +358 2 2318191
mobile: +358 40 6873976
UTU e-mail: marbuc@utu.fi
****************************************************

****************************************************
Marco Bucci, PhD in

Clinical Physiology and Nuclear Medicine