Hi Virendra,
What I don't get is why you would run lGI on the white surface, and specifically on the non topology fixed one? In principle, in mean on the technical side, you could run lGI on the lh.smoothwm which is the corrected surface with no hole and a single component (use mris_compute_lgi --i lh.smoothwm), but biologically I'm not sure it's meaningful. I definitely advise you to look at all coronal sections of the anatomical volume with lh.smoothwm and lh.pial-outer-smoothed / lh.smoothwm-outer-smoothed to get a sense of what you are measuring.
Best,
Marie
On Apr 30, 2014, at 5:39 PM, Virendra Mishra <virendra.mishra@utsouthwestern.edumailto:virendra.mishra@utsouthwestern.edu> wrote:
Hi Marie,
The surface does come after running recon-all but it is not the actual .pial file. It is actually lh.smoothwm.nofix file that I renamed so that I could run IGI.
Thank you for that reply. So, in principle, if I go and find the holes and fill them and make it a continuous surface, then I should be able to compute IGI, right?
Thank you for your help and time.
Sincerely,
Virendra ________________________________ From: Marie Schaer [Marie.Schaer@unige.chmailto:Marie.Schaer@unige.ch] Sent: Wednesday, April 30, 2014 7:02 PM To: Virendra Mishra Cc: freesurfer@nmr.mgh.harvard.edumailto:freesurfer@nmr.mgh.harvard.edu Subject: Re: Question regarding IGI
Hi Virendra,
Where does this surface come from? You didn't get this one after running the default recon-all, right?
Looking at it I noticed a lots of holes and many small unconnected components on the medial aspect. I ran mris_euler_number and got a total of 68 defects, so I'm afraid you'll not be able to run lGI with this surface as the lGI code expect a single continuous surface with no holes. And beside the issue of running lGI until the end, I'm concerned that neither cortical thickness not cortical folding can be reliably estimated based on this surface, and that you'll have trouble registering it to an average template.
Best,
Marie
On Apr 30, 2014, at 3:08 PM, Virendra Mishra <virendra.mishra@utsouthwestern.edumailto:virendra.mishra@utsouthwestern.edu> wrote:
Hello Marie and other freesurfer users,
I am trying to compute IGI for the attached file but at the step of mri_path2label , the code goes into an infinite loop and it is not able to create .label file even for the first .path file. I will be very thankful if you can help me debug this issue.
Thank you,
Sincerely,
Virendra.
________________________________
UT Southwestern Medical Center The future of medicine, today. <lh.pial>
Hi Marie,
Following your advice, I want to compute IGI for pial surface and mris_euler_number gives me 68 holes. What is the procedure to fill these holes?
Thank you,
Sincerely,
Virendra
From: Marie Schaer [mailto:Marie.Schaer@unige.ch] Sent: Wednesday, April 30, 2014 7:50 PM To: Virendra Mishra Cc: freesurfer@nmr.mgh.harvard.edu list Subject: Re: Question regarding IGI
Hi Virendra,
What I don't get is why you would run lGI on the white surface, and specifically on the non topology fixed one? In principle, in mean on the technical side, you could run lGI on the lh.smoothwm which is the corrected surface with no hole and a single component (use mris_compute_lgi --i lh.smoothwm), but biologically I'm not sure it's meaningful. I definitely advise you to look at all coronal sections of the anatomical volume with lh.smoothwm and lh.pial-outer-smoothed / lh.smoothwm-outer-smoothed to get a sense of what you are measuring.
Best,
Marie
On Apr 30, 2014, at 5:39 PM, Virendra Mishra <virendra.mishra@utsouthwestern.edumailto:virendra.mishra@utsouthwestern.edu> wrote:
Hi Marie,
The surface does come after running recon-all but it is not the actual .pial file. It is actually lh.smoothwm.nofix file that I renamed so that I could run IGI.
Thank you for that reply. So, in principle, if I go and find the holes and fill them and make it a continuous surface, then I should be able to compute IGI, right?
Thank you for your help and time.
Sincerely,
Virendra ________________________________ From: Marie Schaer [Marie.Schaer@unige.chmailto:Marie.Schaer@unige.ch] Sent: Wednesday, April 30, 2014 7:02 PM To: Virendra Mishra Cc: freesurfer@nmr.mgh.harvard.edumailto:freesurfer@nmr.mgh.harvard.edu Subject: Re: Question regarding IGI
Hi Virendra,
Where does this surface come from? You didn't get this one after running the default recon-all, right?
Looking at it I noticed a lots of holes and many small unconnected components on the medial aspect. I ran mris_euler_number and got a total of 68 defects, so I'm afraid you'll not be able to run lGI with this surface as the lGI code expect a single continuous surface with no holes. And beside the issue of running lGI until the end, I'm concerned that neither cortical thickness not cortical folding can be reliably estimated based on this surface, and that you'll have trouble registering it to an average template.
Best,
Marie
On Apr 30, 2014, at 3:08 PM, Virendra Mishra <virendra.mishra@utsouthwestern.edumailto:virendra.mishra@utsouthwestern.edu> wrote:
Hello Marie and other freesurfer users,
I am trying to compute IGI for the attached file but at the step of mri_path2label , the code goes into an infinite loop and it is not able to create .label file even for the first .path file. I will be very thankful if you can help me debug this issue.
Thank you,
Sincerely,
Virendra.
________________________________
UT Southwestern Medical Center The future of medicine, today. <lh.pial>
Hi Virendra,
There are tools embedded in the process of surfaces' creation in FreeSurfer which are correcting the topology and prevent the surface to have holes in it. So if you use the pial surface that you get from the recon-all process, you'll be safe for inputing them to the lGI computation.
Best,
Marie
On May 6, 2014, at 2:24 PM, Virendra Mishra <virendra.mishra@utsouthwestern.edumailto:virendra.mishra@utsouthwestern.edu> wrote:
Hi Marie,
Following your advice, I want to compute IGI for pial surface and mris_euler_number gives me 68 holes. What is the procedure to fill these holes?
Thank you,
Sincerely,
Virendra
From: Marie Schaer [mailto:Marie.Schaer@unige.chhttp://unige.ch] Sent: Wednesday, April 30, 2014 7:50 PM To: Virendra Mishra Cc: freesurfer@nmr.mgh.harvard.edumailto:freesurfer@nmr.mgh.harvard.edu list Subject: Re: Question regarding IGI
Hi Virendra,
What I don't get is why you would run lGI on the white surface, and specifically on the non topology fixed one? In principle, in mean on the technical side, you could run lGI on the lh.smoothwm which is the corrected surface with no hole and a single component (use mris_compute_lgi --i lh.smoothwm), but biologically I'm not sure it's meaningful. I definitely advise you to look at all coronal sections of the anatomical volume with lh.smoothwm and lh.pial-outer-smoothed / lh.smoothwm-outer-smoothed to get a sense of what you are measuring.
Best,
Marie
On Apr 30, 2014, at 5:39 PM, Virendra Mishra <virendra.mishra@utsouthwestern.edumailto:virendra.mishra@utsouthwestern.edu> wrote:
Hi Marie,
The surface does come after running recon-all but it is not the actual .pial file. It is actually lh.smoothwm.nofix file that I renamed so that I could run IGI.
Thank you for that reply. So, in principle, if I go and find the holes and fill them and make it a continuous surface, then I should be able to compute IGI, right?
Thank you for your help and time.
Sincerely,
Virendra ________________________________ From: Marie Schaer [Marie.Schaer@unige.chmailto:Marie.Schaer@unige.ch] Sent: Wednesday, April 30, 2014 7:02 PM To: Virendra Mishra Cc: freesurfer@nmr.mgh.harvard.edumailto:freesurfer@nmr.mgh.harvard.edu Subject: Re: Question regarding IGI
Hi Virendra,
Where does this surface come from? You didn't get this one after running the default recon-all, right?
Looking at it I noticed a lots of holes and many small unconnected components on the medial aspect. I ran mris_euler_number and got a total of 68 defects, so I'm afraid you'll not be able to run lGI with this surface as the lGI code expect a single continuous surface with no holes. And beside the issue of running lGI until the end, I'm concerned that neither cortical thickness not cortical folding can be reliably estimated based on this surface, and that you'll have trouble registering it to an average template.
Best,
Marie
On Apr 30, 2014, at 3:08 PM, Virendra Mishra <virendra.mishra@utsouthwestern.edumailto:virendra.mishra@utsouthwestern.edu> wrote:
Hello Marie and other freesurfer users,
I am trying to compute IGI for the attached file but at the step of mri_path2label , the code goes into an infinite loop and it is not able to create .label file even for the first .path file. I will be very thankful if you can help me debug this issue.
Thank you,
Sincerely,
Virendra.
________________________________
UT Southwestern Medical Center The future of medicine, today. <lh.pial>
freesurfer@nmr.mgh.harvard.edu
-
Marie Schaer -
Virendra Mishra