Harvard Oxford Atlas

[ada.sevillamu]

I wanted to add Harvard Oxford atlas, so I used the scritp ea_labeling2atlas in this way: ea_labeling2atlas(‘Harvard Oxford Thr25 2mm Whole Brain (Makris 2006)’)
My problem is that the atlas apppears on the drop-down menu of LEAD DBS, but when I selected it, the visulization 3D does not appear. The Command Windows is always showing these lines:

SPM12: spm_write_sn (v4878) 21:22:21 – 21/02/2017
========================================================================
Completed : 21:22:21 – 21/02/2017

————————————————————————
Running job #1
————————————————————————
Running ‘Image Calculator’

SPM12: spm_imcalc (v6124) 21:22:22 – 21/02/2017
========================================================================
Warning: The images do not all have the same dimensions. – using 1st image.
Warning: The images do not all have same orientation and/or voxel sizes. – using 1st image.
ImCalc Image: C:\Users\ADA\Documents\MATLAB\lead\lead\atlases\Harvard Oxford Thr25 2mm Whole Brain (Makris 2006)\gm_mask.nii
Done ‘Image Calculator’
Done

Do I have to do something else before using the atlas?

[andreashorn]

Hi Ada,

the lines you wrote are just a warning and totally fine. There’s no real “error” in there.
Maybe you just need to wait a bit longer?
When visualizing an atlas for the first time, Lead-DBS needs to build several structures and files first.

Best, Andy

[andreashorn]

Hi Ada,

the lines you wrote are just a warning and totally fine. There’s no real “error” in there.
Maybe you just need to wait a bit longer?
When visualizing an atlas for the first time, Lead-DBS needs to build several structures and files first.

Best, Andy

[ada.sevillamu]

Hi Andy,
you were absolutely right. I have waited a bit longer, and the atlas has worked perfectly.
Thank you.

[mar]

Hi there!
It happens to me exactly the same but then I got an error about memory.
I run >> ea_labeling2atlas(‘Harvard Oxford Thr25 2mm Whole Brain (Makris 2006)’)
and then try to visualize and I get this:

Generating Atlas table (first run with new atlas only). This may take a while…
Building atlas table…

————————————————————————
Running job #1
————————————————————————
Running ‘Image Calculator’

SPM12: spm_imcalc (v6961) 21:26:20 – 03/07/2018
========================================================================
Warning: The images do not all have the same dimensions. – using 1st image.
Warning: The images do not all have same orientation and/or voxel sizes. – using 1st image.
ImCalc Image: C:\Users\Documents\lead\templates\space\MNI_ICBM_2009b_NLIN_ASYM\atlases\Harvard Oxford Thr25 2mm Whole Brain (Makris 2006) – copia\gm_mask.nii
Done ‘Image Calculator’
Done
Error using ea_smoothpatch_inversedistance_double
Requested 18446744069414588570×1 (17179869184.0GB) array exceeds maximum array size preference. Creation of
arrays greater than this limit may take a long time and cause MATLAB to become unresponsive. See array size
limit or preference panel for more information.

if I type memory I get this:
>> memory
Maximum possible array: 10349 MB (1.085e+10 bytes) *
Memory available for all arrays: 10349 MB (1.085e+10 bytes) *
Memory used by MATLAB: 2101 MB (2.203e+09 bytes)
Physical Memory (RAM): 8071 MB (8.463e+09 bytes)

I changed the preferences in matlab to the maximum
But I don´t think I can run it in this laptop
How can that be 17179869184.0GB? Is it normal? or is there something I´m doing wrong?

Thanks a lot!

[andreashorn]

Hi mar,

no clue exactly what happened but it shouldn’t run into array size issues. What you can try:
– Don’t rename the atlas folder to something with a dash (‘ – copia’) but try to avoid blanks and dashes in filepaths where you can
– delete the gm_mask.nii and try again
– check out each nifti image in the generated atlas folder. Any of them looking suspicious? These are just normal nifti files, so you can look at them and compare them to niftis of other atlases.
– if not create the atlas manually (without the labeling2atlas script but e.g. only use selected items from the labeling file that you can e.g. export using the SPM imcalc.

Best, Andy

[mar]

Hi Andy,
I tried what you sugested and same error.
Actually I manually created the atlas using code in labeling2atlas function only selecting the labels I´m interested in: left and right hippocampus. The niftis generated look normal when I display them with spm. Then I tried to visualize the atlas in lead dbs and the same error after a while:

Error using ea_smoothpatch_inversedistance_double
Requested 8774618295792×1 (65376.0GB) array exceeds maximum array size preference. Creation of arrays greater than
this limit may take a long time and cause MATLAB to become unresponsive. See array size limit or preference panel for
more information.

Is it possible to obtain the atlas folder to copy it directly in the atlases folder ready to load it in dbs?
My best,
Mar

[andreashorn]

Mh, weird, I cannot reproduce it here. Maybe a clean install could be helpful.
In any case, made the ready atlas for you: https://www.dropbox.com/s/lc7fx7kijyvs7wo/mar_atlas.zip?dl=0

Best, Andy

[mar]

Thank you very much! Now with your atlas it works! You are so kind.
All the best,
Mar

[Author]

Posts