problem after coregistration

[Nik Haliasos]

Greetings to all,
i am trying to do my first runs with the software. I have one anat.nii and one postop_tra.nii MRI volumes T2 axials
It does not matter what type of coregistration i use, when i go to check the normalization i get a

Skip showing normalization of J:\Medical images\leaddbs\glanat.nii,1
Skip showing normalization of J:\Medical images\leaddbs\glpostop_tra.nii,1
Skip showing normalization of J:\Medical images\leaddbs\glpostop_cor.nii,1

and it does not allow me to view the scans

Then when i go to the next step to reconstruct i get the
Error in ea_checkfiles (line 34)
ea_error(‘Please put a suitable postoperative anatomy file inside patient folder and
normalize it first.’);

any thoughts?

thank you
Nik

• This topic was modified 7 years, 5 months ago by Nik Haliasos.

[ningfei]

Hi Nik,

Would be more helpful if you could show the current contents of your subject folder. Seems you missed some files there.

Best,
Ningfei

[Nik Haliasos]

I did some progress – tried another combination of coregistration and normalization methods and it seems to be working although the registration result is not so great – is there any way to manually correct the registration?

Another issue that came up afterwards is that the pre-recon of the electrodes i cannot get it to work either in the automatic or manual way. I tried focusing on the manual process, the red square and crosshair come up but the slice shown inside this popup windows is the lower postop slice where it does not contain the electrode so matlab afterwards goes into an endless loop of “lowering inital factor to -xxxxxx” and never finds it – i had to Ctrl+C after 1min

Nik

[andreashorn]

Hi Nik,

you should go step by step – if the registration is not good, Lead-DBS will not find the electrode.
We need a bit more info to really help you. You could also check out our Slack channel where you could easily send us some screenshots (See Help/Support above).
It’s not so easy anymore to do normalizations manually, but you could easily perform coregistrations (e.g. post-
to pre or between multiple pre images) manually somewhere else.

In case of post-op CT, you’d need to call the registered CT “rpostop_ct.nii”, in case of postop MRI just call it postop_tra.nii. If you have multiple postop MRIs, you could call them postop_tra.nii and postop_cor.nii, postop_sag.nii.

If all your acquisitions are manually co-registered, just choose “already coregistered” during normalization.

Re normalization, I’d recommend to start with ANTs which seems most robust.

Hope this helps, Andy

[andreashorn]

…one thing that could have been is a segfault error in ANTs, especially likely when running on Windows. This could explain the first error from your first message. Unfortunately, Matlab doesn’t detect the system event segfault and continues as if the called routine had been successful. Ningfei recently fixed the ANTs binaries for Windows, so it could make sense to call “Apply Hotfix” from the Install menu to get up to nightly build state.

[Nik Haliasos]

thank you for your time ..even on a saturday
I am using postop MRI for the moment

will go onto the Slack channel to send some pics if needed

Nik

[Author]

Posts