Coregistration with extra-points on the nose always shows nose points inside the surface

💬 Support & Discussions
1

With our MEGIN system, we always notice the same un-explained behavior. If we take extra digitization points on the nose, they always end-up in a very strange position in on the coregistration GUI.

Example below: the EEG, HPIs, extra head points are OK. But the nose extra head points are inside his sinus.

Screenshot from 2024-08-20 10-06-03
Screenshot from 2024-08-20 10-06-031176×1329 77 KB

Any idea as to what could cause this and what would be the best practice to get a good transformation?

The image was obtained by:

  • running freesurfer with recon-all -all -3T
  • running mne.bem.make_watershed_bem(...)
  • running mne.bem.make_scalp_surfaces(...)
  • running pymeshfix.clean_from_arrays:
    rr, tris = read_surface(subjects_dir / subject / "surf" / "lh.seghead")
    rr, tris = clean_from_arrays(rr, tris)
    write_surface(
        subjects_dir / subject / "surf" / "lh2.seghead", rr, tris, overwrite=True
    )

    rr, tris = read_surface(subjects_dir / subject / "bem" / "outer_skin.surf")
    rr, tris = clean_from_arrays(rr, tris)
    write_surface(
        subjects_dir / subject / "bem" / "outer_skin.surf", rr, tris, overwrite=True
    )
  • running the coregistration GUI, fitting first the fiducials manually placed and then the ICPs.
1 Like