Find Label for MNI coordinate

đź’¬ Support & Discussions
1

Hi,
I have a set of electrode (ECoG) positions in MNI coordinates and a set of anatomical labels (parcellation loaded from FreeSurfer). I couldn’t figure out if and how it is possible to find the most likely label corresponding to each electrode. Is there a function in MNE which just returns the most likely label(s) from the list of labels?

MNE-version: 0.23.dev0
OS: macOS BigSur 11.2.1

import mne
import numpy as np
print(mne.__version__)
subject = 'fsaverage'
subjects_dir = '~/freesurfer_subject'
labels = mne.read_labels_from_annot(subject, parc='aparc',
                                        subjects_dir=subjects_dir,surf_name='pial',hemi='lh')
ch_names = ['test']
ch_coords = [[0,0,0]]
ch_coords =  np.array(ch_coords)/1000
ch_pos = dict(zip(ch_names, ch_coords))
2

Hello @PMM, sorry for the late response. Have you meanwhile been able to resolve this issue?

3

Yes, I wrote my own solution for this. First converting the labels (mne.)vertices_to_mni and then calculating minimal euclidian distances of an electrode position to all labels and returning the labels sorted by distance.