Hi Panos

I think you can trust the white surface (more or less), it's the pial that is tougher. So you might sample say 1mm out from the white and not use the pial at all cheers Bruce On Mon, 30 Jul 2018, Fotiadis, Panagiotis wrote:

Hi Bruce,

Thanks for your prompt response! I see, in that case I will disregard cortical thickness for now. I was also interested in measuring the average signal intensities of the gray and white matter in those ex vivo brains. Do you think it would be prudent to use the resulting surfaces from the ex vivo pipeline (after of course checking their accuracy and making sure the gray/white segmentations look accurate) to measure said intensities, or would you recommend another way?

Thanks, Panos

Panagiotis Fotiadis Senior Imaging Research Technologist J. P. Kistler Stroke Research Center, Department of Neurology Massachusetts General Hospital, Boston, MA T: (617) 643-3869

---

From: freesurfer-bounces@nmr.mgh.harvard.edu freesurfer-bounces@nmr.mgh.harvard.edu on behalf of Bruce Fischl fischl@nmr.mgh.harvard.edu Sent: Monday, July 30, 2018 3:56:33 PM To: Freesurfer support list Subject: Re: [Freesurfer] Ex vivo segmentations   Hi Panos

I wouldn't trust the thickness from the ex vivo pipeline, and in any case those number are way to narrow a range. The normal cortex thickness ranges from about 1.5mm to 4 or 4.5mm. Getting reasonable thickness from ex vivo data is high on our to-do list

cheers Bruce

On Mon, 30 Jul 2018, Fotiadis, Panagiotis wrote:

...

Hi Bruce and Doug,

I have processed a few ex vivo brains with the help of the ex vivo freesurfer pipeline (outlined in https://surfer.nmr.mgh.harvard.edu/fswiki/ExvivoRecon), and have extracted their cortical thickness. So far the range of the cortical thicknesses have been approximately between 3.70-3.90 mm. Does that number sound reasonable for ex vivo brains? I am asking because the

typical

...

cortical thickness that I would expect to find on an in vivo brain would be between 2.05 - 2.35

mm.

...

Thanks in advance for any advice!

Best,

Panos

Panagiotis Fotiadis Senior Imaging Research Technologist J. P. Kistler Stroke Research Center, Department of Neurology Massachusetts General Hospital, Boston, MA T: (617) 643-3869

as I said, I don't think I would trust the ex vivo thicknesses. Definitely not in an absolute sense On Mon, 30 Jul 2018, Fotiadis, Panagiotis wrote:

Hi Bruce,

Thanks, that is actually really helpful.

Also, I realized I misspoke earlier when I was talking about the cortical thickness range. The 3.7-3.9 mm range that I mentioned was referring to global cortical thickness across (not within) the different brains I have processed. The range of cortical thickness within individual brains was indeed wider ( between 2 - 5mm). I was just curious about why the average cortical thickness across brains was higher than what I've seen in the in vivo brains.

Best,

Panos

Panagiotis Fotiadis Senior Imaging Research Technologist J. P. Kistler Stroke Research Center, Department of Neurology Massachusetts General Hospital, Boston, MA T: (617) 643-3869

---

From: freesurfer-bounces@nmr.mgh.harvard.edu freesurfer-bounces@nmr.mgh.harvard.edu on behalf of Bruce Fischl fischl@nmr.mgh.harvard.edu Sent: Monday, July 30, 2018 4:07:48 PM To: Freesurfer support list Subject: Re: [Freesurfer] Ex vivo segmentations   Hi Panos

I think you can trust the white surface (more or less), it's the pial that is tougher. So you might sample say 1mm out from the white and not use the pial at all cheers Bruce On Mon, 30 Jul 2018, Fotiadis, Panagiotis wrote:

...

Hi Bruce,

Thanks for your prompt response! I see, in that case I will disregard cortical thickness for

now. I

...

was also interested in measuring the average signal intensities of the gray and white matter in those ex vivo brains. Do you think it would be prudent to use the resulting surfaces from the ex vivo pipeline (after of course checking their accuracy and making sure the gray/white

segmentations

...

look accurate) to measure said intensities, or would you recommend another way?

Thanks, Panos

Panagiotis Fotiadis Senior Imaging Research Technologist J. P. Kistler Stroke Research Center, Department of Neurology Massachusetts General Hospital, Boston, MA T: (617) 643-3869

---

_

...

From: freesurfer-bounces@nmr.mgh.harvard.edu freesurfer-bounces@nmr.mgh.harvard.edu on behalf of Bruce Fischl fischl@nmr.mgh.harvard.edu Sent: Monday, July 30, 2018 3:56:33 PM To: Freesurfer support list Subject: Re: [Freesurfer] Ex vivo segmentations   Hi Panos

I wouldn't trust the thickness from the ex vivo pipeline, and in any case those number are way to narrow a range. The normal cortex thickness ranges from about 1.5mm to 4 or 4.5mm. Getting reasonable thickness from ex vivo data is high on our to-do list

cheers Bruce

On Mon, 30 Jul 2018, Fotiadis, Panagiotis wrote:

...

Hi Bruce and Doug,

I have processed a few ex vivo brains with the help of the ex vivo freesurfer pipeline (outlined in https://surfer.nmr.mgh.harvard.edu/fswiki/ExvivoRecon), and have extracted their cortical thickness. So far the range of the cortical thicknesses have been approximately between 3.70-3.90 mm. Does that number sound reasonable for ex vivo brains? I am asking because the

typical

...

cortical thickness that I would expect to find on an in vivo brain would be between 2.05 - 2.35

mm.

...

Thanks in advance for any advice!

Best,

Panos

Panagiotis Fotiadis Senior Imaging Research Technologist J. P. Kistler Stroke Research Center, Department of Neurology Massachusetts General Hospital, Boston, MA T: (617) 643-3869

Hi Bruce,

I had a follow-up question concerning measuring the signal intensity of the gray and white matter in those ex vivo brains:

I used mri_vol2surf to create an mgh surface that goes 1mm outside the white matter (to estimate intensity in the gray matter) and another surface that goes 1mm into the white matter (to estimate the signal intensity within the white matter). However, I don't think I can use mri_segstats here to calculate the actual intensity value, as the ex vivo pipeline does not come up with a segmentation option I can use (or at least that is my impression), so I ended up using MATLAB (by MRIreading the mgh file and then calculating a mean of the non-zero array values of the mgh file). Would you recommend a different way to calculate the intensity in those mgh files or do you think that my current MATLAB approach should suffice?

Thanks again for all your help!

Best,

Panos

Panagiotis Fotiadis Senior Imaging Research Technologist J. P. Kistler Stroke Research Center, Department of Neurology Massachusetts General Hospital, Boston, MA T: (617) 643-3869

________________________________ From: freesurfer-bounces@nmr.mgh.harvard.edu freesurfer-bounces@nmr.mgh.harvard.edu on behalf of Fotiadis, Panagiotis PFOTIADIS@mgh.harvard.edu Sent: Monday, July 30, 2018 4:33 PM To: Freesurfer support list Subject: Re: [Freesurfer] Ex vivo segmentations

Cool. Thanks again!

Panagiotis Fotiadis Senior Imaging Research Technologist J. P. Kistler Stroke Research Center, Department of Neurology Massachusetts General Hospital, Boston, MA T: (617) 643-3869 ________________________________ From: freesurfer-bounces@nmr.mgh.harvard.edu freesurfer-bounces@nmr.mgh.harvard.edu on behalf of Bruce Fischl fischl@nmr.mgh.harvard.edu Sent: Monday, July 30, 2018 4:29:35 PM To: Freesurfer support list Subject: Re: [Freesurfer] Ex vivo segmentations

as I said, I don't think I would trust the ex vivo thicknesses. Definitely not in an absolute sense On Mon, 30 Jul 2018, Fotiadis, Panagiotis wrote:

Hi Bruce,

Thanks, that is actually really helpful.

Also, I realized I misspoke earlier when I was talking about the cortical thickness range. The 3.7-3.9 mm range that I mentioned was referring to global cortical thickness across (not within) the different brains I have processed. The range of cortical thickness within individual brains was indeed wider ( between 2 - 5mm). I was just curious about why the average cortical thickness across brains was higher than what I've seen in the in vivo brains.

Best,

Panos

Panagiotis Fotiadis Senior Imaging Research Technologist J. P. Kistler Stroke Research Center, Department of Neurology Massachusetts General Hospital, Boston, MA T: (617) 643-3869

---

From: freesurfer-bounces@nmr.mgh.harvard.edu freesurfer-bounces@nmr.mgh.harvard.edu on behalf of Bruce Fischl fischl@nmr.mgh.harvard.edu Sent: Monday, July 30, 2018 4:07:48 PM To: Freesurfer support list Subject: Re: [Freesurfer] Ex vivo segmentations Hi Panos

I think you can trust the white surface (more or less), it's the pial that is tougher. So you might sample say 1mm out from the white and not use the pial at all cheers Bruce On Mon, 30 Jul 2018, Fotiadis, Panagiotis wrote:

...

Hi Bruce,

Thanks for your prompt response! I see, in that case I will disregard cortical thickness for

now. I

...

was also interested in measuring the average signal intensities of the gray and white matter in those ex vivo brains. Do you think it would be prudent to use the resulting surfaces from the ex vivo pipeline (after of course checking their accuracy and making sure the gray/white

segmentations

...

look accurate) to measure said intensities, or would you recommend another way?

Thanks, Panos

Panagiotis Fotiadis Senior Imaging Research Technologist J. P. Kistler Stroke Research Center, Department of Neurology Massachusetts General Hospital, Boston, MA T: (617) 643-3869

---

_

...

From: freesurfer-bounces@nmr.mgh.harvard.edu freesurfer-bounces@nmr.mgh.harvard.edu on behalf of Bruce Fischl fischl@nmr.mgh.harvard.edu Sent: Monday, July 30, 2018 3:56:33 PM To: Freesurfer support list Subject: Re: [Freesurfer] Ex vivo segmentations Hi Panos

I wouldn't trust the thickness from the ex vivo pipeline, and in any case those number are way to narrow a range. The normal cortex thickness ranges from about 1.5mm to 4 or 4.5mm. Getting reasonable thickness from ex vivo data is high on our to-do list

cheers Bruce

On Mon, 30 Jul 2018, Fotiadis, Panagiotis wrote:

...

Hi Bruce and Doug,

I have processed a few ex vivo brains with the help of the ex vivo freesurfer pipeline (outlined in https://surfer.nmr.mgh.harvard.edu/fswiki/ExvivoRecon), and have extracted their cortical thickness. So far the range of the cortical thicknesses have been approximately between 3.70-3.90 mm. Does that number sound reasonable for ex vivo brains? I am asking because the

typical

...

cortical thickness that I would expect to find on an in vivo brain would be between 2.05 - 2.35

mm.

...

Thanks in advance for any advice!

Best,

Panos

Panagiotis Fotiadis Senior Imaging Research Technologist J. P. Kistler Stroke Research Center, Department of Neurology Massachusetts General Hospital, Boston, MA T: (617) 643-3869

Hi Bruce,

That's actually great for visualizing the new surface as well, thanks! And to calculate the actual intensity at that distance, would I need to resample with mri_vol2surf and then use MATLAB or should I go another way?

Thanks again,

Panos

Panagiotis Fotiadis Senior Imaging Research Technologist J. P. Kistler Stroke Research Center, Department of Neurology Massachusetts General Hospital, Boston, MA T: (617) 643-3869 ________________________________ From: freesurfer-bounces@nmr.mgh.harvard.edu freesurfer-bounces@nmr.mgh.harvard.edu on behalf of Bruce Fischl fischl@nmr.mgh.harvard.edu Sent: Thursday, August 2, 2018 10:24:23 AM To: Freesurfer support list Subject: Re: [Freesurfer] Ex vivo segmentations

Hi Panos

we typically use mri_expand to move the surface inwards and outwards, then sample with the resulting surfaces.

cheers Bruce

On Thu, 2 Aug 2018, Fotiadis, Panagiotis wrote:

Hi Bruce,

I had a follow-up question concerning measuring the signal intensity of the gray and white matter in those ex vivo brains:

I used mri_vol2surf to create an mgh surface that goes 1mm outside the white matter (to estimate intensity in the gray matter) and another surface that goes 1mm into the white matter (to estimate the signal intensity within the white matter). However, I don't think I can use mri_segstats here to calculate the actual intensity value, as the ex vivo pipeline does not come up with a segmentation option I can use (or at least that is my impression), so I ended up using MATLAB (by MRIreading the mgh file and then calculating a mean of the non-zero array values of the mgh file). Would you recommend a different way to calculate the intensity in those mgh files or do you think that my current MATLAB approach should suffice?

Thanks again for all your help!

Best,

Panos

Panagiotis Fotiadis Senior Imaging Research Technologist J. P. Kistler Stroke Research Center, Department of Neurology Massachusetts General Hospital, Boston, MA T: (617) 643-3869

---

From: freesurfer-bounces@nmr.mgh.harvard.edu freesurfer-bounces@nmr.mgh.harvard.edu on behalf of Fotiadis, Panagiotis PFOTIADIS@mgh.harvard.edu Sent: Monday, July 30, 2018 4:33 PM To: Freesurfer support list Subject: Re: [Freesurfer] Ex vivo segmentations

Cool. Thanks again!

Panagiotis Fotiadis Senior Imaging Research Technologist J. P. Kistler Stroke Research Center, Department of Neurology Massachusetts General Hospital, Boston, MA T: (617) 643-3869

---

From: freesurfer-bounces@nmr.mgh.harvard.edu freesurfer-bounces@nmr.mgh.harvard.edu on behalf of Bruce Fischl fischl@nmr.mgh.harvard.edu Sent: Monday, July 30, 2018 4:29:35 PM To: Freesurfer support list Subject: Re: [Freesurfer] Ex vivo segmentations as I said, I don't think I would trust the ex vivo thicknesses. Definitely not in an absolute sense On Mon, 30 Jul 2018, Fotiadis, Panagiotis wrote:

...

Hi Bruce,

Thanks, that is actually really helpful.

Also, I realized I misspoke earlier when I was talking about the cortical thickness range. The 3.7-3.9 mm range that I mentioned was referring to global cortical thickness across (not within)

the

...

different brains I have processed. The range of cortical thickness within individual brains was indeed wider ( between 2 - 5mm). I was just curious about why the average cortical thickness

across

...

brains was higher than what I've seen in the in vivo brains.

Best,

Panos

Panagiotis Fotiadis Senior Imaging Research Technologist J. P. Kistler Stroke Research Center, Department of Neurology Massachusetts General Hospital, Boston, MA T: (617) 643-3869

---

_

...

From: freesurfer-bounces@nmr.mgh.harvard.edu freesurfer-bounces@nmr.mgh.harvard.edu on behalf of Bruce Fischl fischl@nmr.mgh.harvard.edu Sent: Monday, July 30, 2018 4:07:48 PM To: Freesurfer support list Subject: Re: [Freesurfer] Ex vivo segmentations Hi Panos

I think you can trust the white surface (more or less), it's the pial that is tougher. So you might sample say 1mm out from the white and not use the pial at all cheers Bruce On Mon, 30 Jul 2018, Fotiadis, Panagiotis wrote:

...

Hi Bruce,

Thanks for your prompt response! I see, in that case I will disregard cortical thickness for

now. I

...

was also interested in measuring the average signal intensities of the gray and white matter in those ex vivo brains. Do you think it would be prudent to use the resulting surfaces from the ex vivo pipeline (after of course checking their accuracy and making sure the gray/white

segmentations

...

look accurate) to measure said intensities, or would you recommend another way?

Thanks, Panos

Panagiotis Fotiadis Senior Imaging Research Technologist J. P. Kistler Stroke Research Center, Department of Neurology Massachusetts General Hospital, Boston, MA T: (617) 643-3869

---

_

...

_

...

From: freesurfer-bounces@nmr.mgh.harvard.edu freesurfer-bounces@nmr.mgh.harvard.edu on behalf

of

...

...

Bruce Fischl fischl@nmr.mgh.harvard.edu Sent: Monday, July 30, 2018 3:56:33 PM To: Freesurfer support list Subject: Re: [Freesurfer] Ex vivo segmentations Hi Panos

I wouldn't trust the thickness from the ex vivo pipeline, and in any case those number are way to narrow a range. The normal cortex thickness ranges from about 1.5mm to 4 or 4.5mm. Getting reasonable thickness from ex vivo data is high on our to-do list

cheers Bruce

On Mon, 30 Jul 2018, Fotiadis, Panagiotis wrote:

...

Hi Bruce and Doug,

I have processed a few ex vivo brains with the help of the ex vivo freesurfer pipeline

(outlined

...

...

...

in https://surfer.nmr.mgh.harvard.edu/fswiki/ExvivoRecon), and have extracted their cortical thickness. So far the range of the cortical thicknesses have been approximately between 3.70-3.90 mm. Does that number sound reasonable for ex vivo brains? I am asking because the

typical

...

cortical thickness that I would expect to find on an in vivo brain would be between 2.05 -

2.35

...

...

mm.

...

Thanks in advance for any advice!

Best,

Panos

Panagiotis Fotiadis Senior Imaging Research Technologist J. P. Kistler Stroke Research Center, Department of Neurology Massachusetts General Hospital, Boston, MA T: (617) 643-3869

Awesome, thanks Bruce!

Panagiotis Fotiadis Senior Imaging Research Technologist J. P. Kistler Stroke Research Center, Department of Neurology Massachusetts General Hospital, Boston, MA T: (617) 643-3869 ________________________________ From: freesurfer-bounces@nmr.mgh.harvard.edu freesurfer-bounces@nmr.mgh.harvard.edu on behalf of Bruce Fischl fischl@nmr.mgh.harvard.edu Sent: Thursday, August 2, 2018 11:13:02 AM To: Freesurfer support list Subject: Re: [Freesurfer] Ex vivo segmentations

yes, use vol2surf. I'm not sure what you want to do with the results, but certainly matlab will give you the max flexibility On Thu, 2 Aug 2018, Fotiadis, Panagiotis wrote:

Hi Bruce,

That's actually great for visualizing the new surface as well, thanks! And to calculate the actual intensity at that distance, would I need to resample with mri_vol2surf and then use MATLAB or should I go another way?

Thanks again,

Panos

Panagiotis Fotiadis Senior Imaging Research Technologist J. P. Kistler Stroke Research Center, Department of Neurology Massachusetts General Hospital, Boston, MA T: (617) 643-3869

---

From: freesurfer-bounces@nmr.mgh.harvard.edu freesurfer-bounces@nmr.mgh.harvard.edu on behalf of Bruce Fischl fischl@nmr.mgh.harvard.edu Sent: Thursday, August 2, 2018 10:24:23 AM To: Freesurfer support list Subject: Re: [Freesurfer] Ex vivo segmentations Hi Panos

we typically use mri_expand to move the surface inwards and outwards, then sample with the resulting surfaces.

cheers Bruce

On Thu, 2 Aug 2018, Fotiadis, Panagiotis wrote:

...

Hi Bruce,

I had a follow-up question concerning measuring the signal intensity of the gray and white

matter in

...

those ex vivo brains:

I used mri_vol2surf to create an mgh surface that goes 1mm outside the white matter (to estimate intensity in the gray matter) and another surface that goes 1mm into the white matter (to estimate the signal intensity within the white matter). However, I don't think I can use mri_segstats here

to

...

calculate the actual intensity value, as the ex vivo pipeline does not come up with a segmentation option I can use (or at least that is my impression), so I ended up using MATLAB (by MRIreading

the

...

mgh file and then calculating a mean of the non-zero array values of the mgh file). Would you recommend a different way to calculate the intensity in those mgh files or do you think that my current MATLAB approach should suffice?

Thanks again for all your help!

Best,

Panos

Panagiotis Fotiadis Senior Imaging Research Technologist J. P. Kistler Stroke Research Center, Department of Neurology Massachusetts General Hospital, Boston, MA T: (617) 643-3869

---

_

...

From: freesurfer-bounces@nmr.mgh.harvard.edu freesurfer-bounces@nmr.mgh.harvard.edu on behalf of Fotiadis, Panagiotis PFOTIADIS@mgh.harvard.edu Sent: Monday, July 30, 2018 4:33 PM To: Freesurfer support list Subject: Re: [Freesurfer] Ex vivo segmentations

Cool. Thanks again!

Panagiotis Fotiadis Senior Imaging Research Technologist J. P. Kistler Stroke Research Center, Department of Neurology Massachusetts General Hospital, Boston, MA T: (617) 643-3869

---

_

...

From: freesurfer-bounces@nmr.mgh.harvard.edu freesurfer-bounces@nmr.mgh.harvard.edu on behalf of Bruce Fischl fischl@nmr.mgh.harvard.edu Sent: Monday, July 30, 2018 4:29:35 PM To: Freesurfer support list Subject: Re: [Freesurfer] Ex vivo segmentations as I said, I don't think I would trust the ex vivo thicknesses. Definitely not in an absolute sense On Mon, 30 Jul 2018, Fotiadis, Panagiotis wrote:

...

Hi Bruce,

Thanks, that is actually really helpful.

Also, I realized I misspoke earlier when I was talking about the cortical thickness range. The 3.7-3.9 mm range that I mentioned was referring to global cortical thickness across (not within)

the

...

different brains I have processed. The range of cortical thickness within individual brains was indeed wider ( between 2 - 5mm). I was just curious about why the average cortical thickness

across

...

brains was higher than what I've seen in the in vivo brains.

Best,

Panos

Panagiotis Fotiadis Senior Imaging Research Technologist J. P. Kistler Stroke Research Center, Department of Neurology Massachusetts General Hospital, Boston, MA T: (617) 643-3869

---

_

...

_

...

From: freesurfer-bounces@nmr.mgh.harvard.edu freesurfer-bounces@nmr.mgh.harvard.edu on behalf

of

...

...

Bruce Fischl fischl@nmr.mgh.harvard.edu Sent: Monday, July 30, 2018 4:07:48 PM To: Freesurfer support list Subject: Re: [Freesurfer] Ex vivo segmentations Hi Panos

I think you can trust the white surface (more or less), it's the pial that is tougher. So you might sample say 1mm out from the white and not use the pial at all cheers Bruce On Mon, 30 Jul 2018, Fotiadis, Panagiotis wrote:

...

Hi Bruce,

Thanks for your prompt response! I see, in that case I will disregard cortical thickness for

now. I

...

was also interested in measuring the average signal intensities of the gray and white matter

in

...

...

...

those ex vivo brains. Do you think it would be prudent to use the resulting surfaces from the

ex

...

...

...

vivo pipeline (after of course checking their accuracy and making sure the gray/white

segmentations

...

look accurate) to measure said intensities, or would you recommend another way?

Thanks, Panos

Panagiotis Fotiadis Senior Imaging Research Technologist J. P. Kistler Stroke Research Center, Department of Neurology Massachusetts General Hospital, Boston, MA T: (617) 643-3869

---

_

...

_

...

_

...

From: freesurfer-bounces@nmr.mgh.harvard.edu freesurfer-bounces@nmr.mgh.harvard.edu on

behalf

...

of

...

...

Bruce Fischl fischl@nmr.mgh.harvard.edu Sent: Monday, July 30, 2018 3:56:33 PM To: Freesurfer support list Subject: Re: [Freesurfer] Ex vivo segmentations Hi Panos

I wouldn't trust the thickness from the ex vivo pipeline, and in any case those number are way to narrow a range. The normal cortex thickness ranges from about 1.5mm to 4 or 4.5mm. Getting reasonable thickness from ex vivo data is high on our to-do list

cheers Bruce

On Mon, 30 Jul 2018, Fotiadis, Panagiotis wrote:

...

Hi Bruce and Doug,

I have processed a few ex vivo brains with the help of the ex vivo freesurfer pipeline

(outlined

...

...

...

in https://surfer.nmr.mgh.harvard.edu/fswiki/ExvivoRecon), and have extracted their cortical thickness. So far the range of the cortical thicknesses have been approximately between 3.70-3.90 mm. Does that number sound reasonable for ex vivo brains? I am asking because the

typical

...

cortical thickness that I would expect to find on an in vivo brain would be between 2.05 -

2.35

...

...

mm.

...

Thanks in advance for any advice!

Best,

Panos

Panagiotis Fotiadis Senior Imaging Research Technologist J. P. Kistler Stroke Research Center, Department of Neurology Massachusetts General Hospital, Boston, MA T: (617) 643-3869