Error When Removing Erroneous Channels

[zborden3]

Hello,

We hope you are doing well. My colleague and I are working on a project that uses the get_chanlocs plugin to localize scalp electrodes from 3D scans. We have been following your protocol exclusively from SCNN get_chanlocs user guide, and we have been able to capture adequate scans using the Structure Sensor Pro and ItSeez3D. We have successfully uploaded the 3D model to EEGLAB and get_chanlocs. Additionally, we have been able to properly locate the fiducial points and manually mark the location of each electrode on our 3D model. We have also been using an MNI model template within the get_chanlocs program.

We are encountering an issue when we attempt to take the next step to remove and reselect erroneous electrode selections and pairings. We are able to properly remove the pairing of interest, through selecting the channel and pressing “r” from the Current Model Locations. However, when we are prompted to select new location pairings, through pressing “s", we are unable to select a new location. The window that pops up shows a large red blob over the person's head, making it impossible to select a new location for the erroneous electrode channel.

We initially reached out to you and determined that the issue may arise from a scaling issue from cm to mm and that the maker size was out of proportion. We tried contacting ItSeez3D Support and they sent us resized scans in mm and this still did not solve the issue as it was almost impossible to select the fiducial points accurately each time. We also tried redownloaded the master branch to the most recent version, and we replaced the directory that we were using with the contents of the master branch zip file and this did not solve the issue either.

We are wondering how we can select a more accurate location than our original erroneous placement, given that the pop-up window shows a large red blob that blocks our ability to select a better location? What are the steps that we should be taking in order to select new locations when given this pop-up window?

Thanks so much!

[neuromechanist]

We talked about the problem. Even when asking ItSeez3D and converting the scan to mm, the scaling issue still remains.

@zborden3, please test the master branch and check if the scaling fix I added before could resolve the issue. Please try both scans (one corrected by ItSeez3D and one without their correction) and let me know if you can get the correct digitization.

[zborden3]

Hello, thank you very much for your reply. I have re-downloaded the master branch and tried both scans (uncorrected and corrected) to see if I could get the correct digitization as you suggested. The scans that we received from ItSeez3D, which were corrected from a cm scale to a mm scale, seemed to solve the issue when we re-downloaded and re-tested both the master branch and the newUI as we got the correct digitization here. However, our original scans, which were not corrected and possibly still in a cm scale, did not resolve the issue and resulted in the same problem seen above. While this is definitely good news that we have a potential solution, this still presents a problem for us with moving forward with our research as the maker size and proportions of our scan will always need to be corrected each time we scan a new participant. We are unsure of how to change the maker size and the scaling/proportion of our scan so that it is always in mm, so we will have to contact ItSeez3D Support each time we scan someone to ensure that the maker size is correct and have them resize our scan to a mm scale. How would you suggest proceeding from here so that we can ensure that we always have corrected scans that will yield the correct digitization? Thanks!

[zborden3]

Here is a picture of the correct digitization seen when we used the corrected scans from ItSeez3D in mm.

[neuromechanist]

Thanks, @zborden3, for the update.

For ItSeez3D, I think you could ask them to give you a script that converts the scales from mm to cm. A more permanent fix for them would be adding an option to determine the dimension before export.

The other point is to try to only include the head in the capture volume. I assume the problem arises because you are potentially including a very large capture volume, including the shoulders and torso, so the ItSeez3D app falls to the cm scale.

Last, please send the corrected and uncorrected scan to my email. I want to investigate why the fix I implemented did not address account for your issue.