Dear Freesurfer experts,
We are currently working with some scans that were collected on a GE 3T scanner. The goal is to analyze cortical thickness. The scans seem to be relatively high quality with the exception that there are what appear to be inhomogeneities in the intensity of the gray matter (appears too bright in places), which is causing the WM to extend/jump into cortex (see attached image). We're wondering if there are any parameter adjustments to recon that might help with this as the errors occur frequently enough that fixing them all manually will be fairly cumbersome. This type of error doesn’t seem to be discussed very much in the available information on freesurfer quality control as far as we’ve seen. Thanks in advance for any advice you might be able to offer.
Best regards, Ben
[image: Inline image 1]
Hello Ben,
I am no expert, I can only talk from my experience. I've had similar images in my data. Those might be blood vessels or dura that were included in the surface. If they are included in the surface on more than 5 slices and pretty well visible you could delete those voxels from the brainmask. From the looks of the attached screenshots however I wouldn't worry about the brainmask but rather remove a few voxels from the wm.mgz. Then you rerun the subject with the -wm flag and check again. If it is only on one or two slices, though, I wouldn't edit it at all. The surfaces are just not going to be perfect.
Hope I could help. Clara
----- Ursprüngliche Mail ----- Von: "Benjamin Baird" benjamin.s.baird@gmail.com An: freesurfer@nmr.mgh.harvard.edu Gesendet: Donnerstag, 1. September 2016 21:56:58 Betreff: [Freesurfer] white matter errors due to inhomogeneities (bright spots) in cortex
Dear Freesurfer experts,
We are currently working with some scans that were collected on a GE 3T scanner. The goal is to analyze cortical thickness. The scans seem to be relatively high quality with the exception that there are what appear to be inhomogeneities in the intensity of the gray matter (appears too bright in places), which is causing the WM to extend/jump into cortex (see attached image). We're wondering if there are any parameter adjustments to recon that might help with this as the errors occur frequently enough that fixing them all manually will be fairly cumbersome. This type of error doesn’t seem to be discussed very much in the available information on freesurfer quality control as far as we’ve seen. Thanks in advance for any advice you might be able to offer.
Best regards, Ben
_______________________________________________ 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.
Hi Bruce and Clara,
Thanks for your comments. Indeed the bright points are segmented as WM in wm.mgz, however these errors do not occur only in a few places but in many slices throughout the brain so manual edits will be quite extensive, which is why we wanted to double check that there is no parameter tweak to recon that might at least help some of these errors. But sounds like there's no way to adjust the parameters of recon to mitigate this type of error? If not we will move forward with manual edits to wm.mgz as you suggest. Thanks!
Best regards, Ben
On Fri, Sep 2, 2016 at 8:36 AM, Clara Kühn ckuehn@cbs.mpg.de wrote:
Hello Ben,
I am no expert, I can only talk from my experience. I've had similar images in my data. Those might be blood vessels or dura that were included in the surface. If they are included in the surface on more than 5 slices and pretty well visible you could delete those voxels from the brainmask. From the looks of the attached screenshots however I wouldn't worry about the brainmask but rather remove a few voxels from the wm.mgz. Then you rerun the subject with the -wm flag and check again. If it is only on one or two slices, though, I wouldn't edit it at all. The surfaces are just not going to be perfect.
Hope I could help. Clara
----- Ursprüngliche Mail ----- Von: "Benjamin Baird" benjamin.s.baird@gmail.com An: freesurfer@nmr.mgh.harvard.edu Gesendet: Donnerstag, 1. September 2016 21:56:58 Betreff: [Freesurfer] white matter errors due to inhomogeneities (bright spots) in cortex
Dear Freesurfer experts,
We are currently working with some scans that were collected on a GE 3T scanner. The goal is to analyze cortical thickness. The scans seem to be relatively high quality with the exception that there are what appear to be inhomogeneities in the intensity of the gray matter (appears too bright in places), which is causing the WM to extend/jump into cortex (see attached image). We're wondering if there are any parameter adjustments to recon that might help with this as the errors occur frequently enough that fixing them all manually will be fairly cumbersome. This type of error doesn’t seem to be discussed very much in the available information on freesurfer quality control as far as we’ve seen. Thanks in advance for any advice you might be able to offer.
Best regards, Ben
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.
Hi Bruce and Clara,
Many thanks for your comments. Indeed the bright points are segmented as WM in wm.mgz, however these errors do not occur only in a few places but in many slices throughout the brain so manual edits will be very extensive, which is why we wanted to double check that there is no parameter tweak to recon that might at least help some of these errors. But sounds like there's no way to adjust the parameters of recon to mitigate this type of error? If not we will move forward with manual edits to wm.mgz as you suggest. Also I'm wondering if this is a common issue with GE scanners, as I've never encountered it before on Siemens, and I recently saw a post to the FS group from someone else encountering the same issue on GE
Best regards, Ben
On Fri, Sep 2, 2016 at 8:36 AM, Clara Kühn ckuehn@cbs.mpg.de wrote:
Hello Ben,
I am no expert, I can only talk from my experience. I've had similar images in my data. Those might be blood vessels or dura that were included in the surface. If they are included in the surface on more than 5 slices and pretty well visible you could delete those voxels from the brainmask. From the looks of the attached screenshots however I wouldn't worry about the brainmask but rather remove a few voxels from the wm.mgz. Then you rerun the subject with the -wm flag and check again. If it is only on one or two slices, though, I wouldn't edit it at all. The surfaces are just not going to be perfect.
Hope I could help. Clara
----- Ursprüngliche Mail ----- Von: "Benjamin Baird" benjamin.s.baird@gmail.com An: freesurfer@nmr.mgh.harvard.edu Gesendet: Donnerstag, 1. September 2016 21:56:58 Betreff: [Freesurfer] white matter errors due to inhomogeneities (bright spots) in cortex
Dear Freesurfer experts,
We are currently working with some scans that were collected on a GE 3T scanner. The goal is to analyze cortical thickness. The scans seem to be relatively high quality with the exception that there are what appear to be inhomogeneities in the intensity of the gray matter (appears too bright in places), which is causing the WM to extend/jump into cortex (see attached image). We're wondering if there are any parameter adjustments to recon that might help with this as the errors occur frequently enough that fixing them all manually will be fairly cumbersome. This type of error doesn’t seem to be discussed very much in the available information on freesurfer quality control as far as we’ve seen. Thanks in advance for any advice you might be able to offer.
Best regards, Ben
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.
Hi Ben
sorry, but I don't know how to fix it automatically. You have voxels that look just like wm in the gm - that's going to be tough to avoid. You could try messing with some of the mris_make_surfaces options, but I don't have much confidence that it will help.
Any idea why you are getting the bright voxels in the first place?
Bruce
On Mon, 5 Sep 2016, Benjamin Baird wrote:
Hi Bruce and Clara, Many thanks for your comments. Indeed the bright points are segmented as WM in wm.mgz, however these errors do not occur only in a few places but in many slices throughout the brain so manual edits will be very extensive, which is why we wanted to double check that there is no parameter tweak to recon that might at least help some of these errors. But sounds like there's no way to adjust the parameters of recon to mitigate this type of error? If not we will move forward with manual edits to wm.mgz as you suggest. Also I'm wondering if this is a common issue with GE scanners, as I've never encountered it before on Siemens, and I recently saw a post to the FS group from someone else encountering the same issue on GE
Best regards, Ben
On Fri, Sep 2, 2016 at 8:36 AM, Clara Kühn ckuehn@cbs.mpg.de wrote: Hello Ben,
I am no expert, I can only talk from my experience. I've had similar images in my data. Those might be blood vessels or dura that were included in the surface. If they are included in the surface on more than 5 slices and pretty well visible you could delete those voxels from the brainmask. From the looks of the attached screenshots however I wouldn't worry about the brainmask but rather remove a few voxels from the wm.mgz. Then you rerun the subject with the -wm flag and check again. If it is only on one or two slices, though, I wouldn't edit it at all. The surfaces are just not going to be perfect. Hope I could help. Clara ----- Ursprüngliche Mail ----- Von: "Benjamin Baird" <benjamin.s.baird@gmail.com> An: freesurfer@nmr.mgh.harvard.edu Gesendet: Donnerstag, 1. September 2016 21:56:58 Betreff: [Freesurfer] white matter errors due to inhomogeneities (bright spots) in cortex Dear Freesurfer experts, We are currently working with some scans that were collected on a GE 3T scanner. The goal is to analyze cortical thickness. The scans seem to be relatively high quality with the exception that there are what appear to be inhomogeneities in the intensity of the gray matter (appears too bright in places), which is causing the WM to extend/jump into cortex (see attached image). We're wondering if there are any parameter adjustments to recon that might help with this as the errors occur frequently enough that fixing them all manually will be fairly cumbersome. This type of error doesn’t seem to be discussed very much in the available information on freesurfer quality control as far as we’ve seen. Thanks in advance for any advice you might be able to offer. Best regards, Ben _______________________________________________ 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.
Not sure it's helpful.
The simplest way I can think of is to directly edit WM.mgz. You can load wm.mgz into matlab, it's basically a 3D matrix. You can directly set values for wm matrix, for instance
wm(wm(:)>200)=110;
Then you save the new wm matrix into wm.mgz.
On 9/5/16 10:54 AM, Benjamin Baird wrote:
Hi Bruce and Clara,
Many thanks for your comments. Indeed the bright points are segmented as WM in wm.mgz, however these errors do not occur only in a few places but in many slices throughout the brain so manual edits will be very extensive, which is why we wanted to double check that there is no parameter tweak to recon that might at least help some of these errors. But sounds like there's no way to adjust the parameters of recon to mitigate this type of error? If not we will move forward with manual edits to wm.mgz as you suggest. Also I'm wondering if this is a common issue with GE scanners, as I've never encountered it before on Siemens, and I recently saw a post to the FS group from someone else encountering the same issue on GE
Best regards, Ben
On Fri, Sep 2, 2016 at 8:36 AM, Clara Kühn <ckuehn@cbs.mpg.de mailto:ckuehn@cbs.mpg.de> wrote:
Hello Ben, I am no expert, I can only talk from my experience. I've had similar images in my data. Those might be blood vessels or dura that were included in the surface. If they are included in the surface on more than 5 slices and pretty well visible you could delete those voxels from the brainmask. From the looks of the attached screenshots however I wouldn't worry about the brainmask but rather remove a few voxels from the wm.mgz. Then you rerun the subject with the -wm flag and check again. If it is only on one or two slices, though, I wouldn't edit it at all. The surfaces are just not going to be perfect. Hope I could help. Clara ----- Ursprüngliche Mail ----- Von: "Benjamin Baird" <benjamin.s.baird@gmail.com <mailto:benjamin.s.baird@gmail.com>> An: freesurfer@nmr.mgh.harvard.edu <mailto:freesurfer@nmr.mgh.harvard.edu> Gesendet: Donnerstag, 1. September 2016 21:56:58 Betreff: [Freesurfer] white matter errors due to inhomogeneities (bright spots) in cortex Dear Freesurfer experts, We are currently working with some scans that were collected on a GE 3T scanner. The goal is to analyze cortical thickness. The scans seem to be relatively high quality with the exception that there are what appear to be inhomogeneities in the intensity of the gray matter (appears too bright in places), which is causing the WM to extend/jump into cortex (see attached image). We're wondering if there are any parameter adjustments to recon that might help with this as the errors occur frequently enough that fixing them all manually will be fairly cumbersome. This type of error doesn’t seem to be discussed very much in the available information on freesurfer quality control as far as we’ve seen. Thanks in advance for any advice you might be able to offer. Best regards, Ben _______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu <mailto:Freesurfer@nmr.mgh.harvard.edu> https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer <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 <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 <mailto:Freesurfer@nmr.mgh.harvard.edu> https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer <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 <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@nmr.mgh.harvard.edu
-
Benjamin Baird -
Benjamin Baird -
Bruce Fischl -
Clara Kühn -
Ruyuan Zhang