Voronoi and head model error in workshop 2020

[AlexBo924]

Good afternoon Thomas/Zhengchen and Edouard,

I recently trying testing again nirstorm and have a few issues/questions.

First, when trying to calculate Voronoi and head model in the cortical reconstruction demo provided in the workshop 2020, I face 2 errors.

These errors are not present in the demo from the 2018 workshop (ONLY in the demo provided in 2020). I dug a bit and realized that in the 2018 demo we have the 5 tissues segmentation, but not in the 2020 workshop (present but when you view it there is nothing.

If you could please confirm if that might explain why I get this error in the demo from 2020 but not 2018.

Also, while I was able to calculate the voronoi and head model in the 2018 demo, the size of the voronoi and head model were quite different from the one present originally in the demo. I was still able to extract the sensitivities and perform a simple cortical reconstruction, but again is it normal to have different sizes despite the same data (visual task provided in the demo)? I included in a zip file the old and new voronoi and head model

subjectimage_T1_tess_cortex_mid_high_25000V_voronoi_ORIGINAL / 1.5 MB subjectimage_T1_tess_cortex_mid_high_25000V_voronoi_RECALCULATED / 1.2 MB headmodel_nirs_mcx_fluence_ORIGINAL / 7 MB headmodel_nirs_mcx_fluence_RECALCULATED / 0.2 MB

Size_diff.zip

Thanks again

Voronoi error

Head model error

5tissue segmentation from the 2018 workshop

5tissue segmentation from the 2020 workshop

voronoi from the 2018 workshop / ORIGINAL

voronoi recalculated from the 2018 demo / RECALCULATED

Thank you again all in advance

[Edouard2laire]

it's weird. Can it be just a downloading issue ? Where did you get the data from ? because we had no problem during the workshop

[AlexBo924]

Hello Edouard, sorry for the late response.

I got the data from the download link provided during the workshop. I doubled checked again (the original zip file, and indeed the 5tissues segmentation (size and original matlab file are very different / smaller) than the 5tissues segmentation from the workshop back in 2018.

When you open the 5tissues segmentation from the 2020 workshop in brainstorm you can't see anything (compared to the one from the 2018 workshop).

The reason (I think) why we didn't have an issue during the workshop is because the head model, sensitives and fluences were either already provided or already pre-calculated.

If you can maybe try to open the 5tissues segmentation or to re-calculate voronoi (both for the workshop 2020) and see if you get the same error (of course whenever you have the time).

I think probably the original 5tissues

segmentation (good one) was either missing from the demo or not calculated correctly.

Its just for your information/future workshop.

Thank you Edouard.

subjectimage_191024_1218.zip

[Edouard2laire]

Hey. I think I know what the problem is :) So when you double click on the segmentation file, the display is not good and it's actually doesn't display anything. As you can see the colorer goes from 0 to 1, or the segmentation values are from 0 to 5 so you can't see anything. To see the segmentation file, you need to right-click and use overlay on MRI viewer.

Here is what I have from the tutorial for the segmentation (top) and Voronoi (bottom)

[AlexBo924]

Hi Edouard,

Indeed that was the issue (I don't know why there is this difference between the 2 workshops, but it's ok).

Also, just on the side, while we do see the segmentation and Voronoi, trying to re-calculate the Voronoi or head model (after deleting them first) for the 2020 workshop demo, we still face the error mentioned above (but again it's ok, it does work in general, as well as for the 2018 demo).

Thank you again, Edouard, for looking into it, I really appreciate it.