External Email - Use Caution

Hi Doug,

The relevant wiki page is: https://surfer.nmr.mgh.harvard.edu/fswiki/LabelsClutsAnnotationFiles#Annotat...

Chris

________________________________ From: freesurfer-bounces@nmr.mgh.harvard.edu freesurfer-bounces@nmr.mgh.harvard.edu on behalf of Greve, Douglas N.,Ph.D. DGREVE@mgh.harvard.edu Sent: Friday, September 28, 2018 12:47:36 PM To: freesurfer@nmr.mgh.harvard.edu Subject: Re: [Freesurfer] Clarification about .annot file format

Thanks Chris, can you point us to the offending wiki page? doug

On 9/28/18 12:44 PM, Christopher Markiewicz wrote:

External Email - Use Caution

Hi all,

I believe I've gotten to the bottom of this, and the problem is incorrect documentation, both in the read/write_annotation.m files and in the wiki.

Taking the FreeSurfer C source as the de facto specification, we can refer to the colortab.c file. Here, the labels are not constructed or decoded using the transparency (255 - alpha) information at all, but are simply (R + G << 8 + B << 16).

For more complete discussion, see:

https://github.com/nipy/nibabel/issues/649#issuecomment-424755969

I think the most parsimonious resolution of this confusion is simply to update the documentation on the wiki page (I cannot; someone with permissions will need to), and to patch the MATLAB code to build its labels from RGB values alone. I will submit a patch to the GitHub branch.

Please let me know if I've gotten something wrong.

Best, Chris Markiewicz

________________________________ From: freesurfer-bounces@nmr.mgh.harvard.edumailto:freesurfer-bounces@nmr.mgh.harvard.edu freesurfer-bounces@nmr.mgh.harvard.edumailto:freesurfer-bounces@nmr.mgh.harvard.edu on behalf of paul mccarthy pauldmccarthy@gmail.commailto:pauldmccarthy@gmail.com Sent: Thursday, September 20, 2018 11:39:53 AM To: freesurfer@nmr.mgh.harvard.edumailto:freesurfer@nmr.mgh.harvard.edu Subject: [Freesurfer] Clarification about .annot file format

External Email - Use Caution

Howdy,

There is a discussion regarding the .annot file format over at the nibabel github repository, regarding the interpretation of the flag/alpha value:

https://github.com/nipy/nibabel/issues/649

If one wishes to create an .annot file with colours that have a transparency value, how should it be stored in the file? By piecing together the information from the freesurfer code [1,2], and Graham Wideman's helpful notes [3], I came to the conclusion that alpha values should be stored as 1-the actual value. But it would be great to get some clarification on this from somebody who knows more than me

Thanks very much,

Paul McCarthy

[1] https://github.com/freesurfer/freesurfer/blob/dev/matlab/read_annotation.m [2] https://github.com/freesurfer/freesurfer/blob/dev/matlab/write_annotation.m [3] https://surfer.nmr.mgh.harvard.edu/fswiki/LabelsClutsAnnotationFiles#Annotat...

_______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edumailto:Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer

External Email - Use Caution

Doug,

I think that's the only one.

Chris

________________________________ From: freesurfer-bounces@nmr.mgh.harvard.edu freesurfer-bounces@nmr.mgh.harvard.edu on behalf of Greve, Douglas N.,Ph.D. DGREVE@mgh.harvard.edu Sent: Friday, September 28, 2018 1:09:50 PM To: freesurfer@nmr.mgh.harvard.edu Subject: Re: [Freesurfer] Clarification about .annot file format

Is this the offending line? or is there more? value = (A * 2563) + (B * 2562) + (G * 256) + (R)

On 9/28/18 1:03 PM, Christopher Markiewicz wrote:

External Email - Use Caution

Hi Doug,

The relevant wiki page is: https://surfer.nmr.mgh.harvard.edu/fswiki/LabelsClutsAnnotationFiles#Annotat...

Chris

________________________________ From: freesurfer-bounces@nmr.mgh.harvard.edumailto:freesurfer-bounces@nmr.mgh.harvard.edu freesurfer-bounces@nmr.mgh.harvard.edumailto:freesurfer-bounces@nmr.mgh.harvard.edu on behalf of Greve, Douglas N.,Ph.D. DGREVE@mgh.harvard.edumailto:DGREVE@mgh.harvard.edu Sent: Friday, September 28, 2018 12:47:36 PM To: freesurfer@nmr.mgh.harvard.edumailto:freesurfer@nmr.mgh.harvard.edu Subject: Re: [Freesurfer] Clarification about .annot file format

Thanks Chris, can you point us to the offending wiki page? doug

On 9/28/18 12:44 PM, Christopher Markiewicz wrote:

External Email - Use Caution

Hi all,

I believe I've gotten to the bottom of this, and the problem is incorrect documentation, both in the read/write_annotation.m files and in the wiki.

Taking the FreeSurfer C source as the de facto specification, we can refer to the colortab.c file. Here, the labels are not constructed or decoded using the transparency (255 - alpha) information at all, but are simply (R + G << 8 + B << 16).

For more complete discussion, see:

https://github.com/nipy/nibabel/issues/649#issuecomment-424755969

I think the most parsimonious resolution of this confusion is simply to update the documentation on the wiki page (I cannot; someone with permissions will need to), and to patch the MATLAB code to build its labels from RGB values alone. I will submit a patch to the GitHub branch.

Please let me know if I've gotten something wrong.

Best, Chris Markiewicz

________________________________ From: freesurfer-bounces@nmr.mgh.harvard.edumailto:freesurfer-bounces@nmr.mgh.harvard.edu freesurfer-bounces@nmr.mgh.harvard.edumailto:freesurfer-bounces@nmr.mgh.harvard.edu on behalf of paul mccarthy pauldmccarthy@gmail.commailto:pauldmccarthy@gmail.com Sent: Thursday, September 20, 2018 11:39:53 AM To: freesurfer@nmr.mgh.harvard.edumailto:freesurfer@nmr.mgh.harvard.edu Subject: [Freesurfer] Clarification about .annot file format

External Email - Use Caution

Howdy,

There is a discussion regarding the .annot file format over at the nibabel github repository, regarding the interpretation of the flag/alpha value:

https://github.com/nipy/nibabel/issues/649

If one wishes to create an .annot file with colours that have a transparency value, how should it be stored in the file? By piecing together the information from the freesurfer code [1,2], and Graham Wideman's helpful notes [3], I came to the conclusion that alpha values should be stored as 1-the actual value. But it would be great to get some clarification on this from somebody who knows more than me

Thanks very much,

Paul McCarthy

[1] https://github.com/freesurfer/freesurfer/blob/dev/matlab/read_annotation.m [2] https://github.com/freesurfer/freesurfer/blob/dev/matlab/write_annotation.m [3] https://surfer.nmr.mgh.harvard.edu/fswiki/LabelsClutsAnnotationFiles#Annotat...

_______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edumailto:Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer

_______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edumailto:Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer