Note
Go to the end to download the full example code.
Plot Intracranial Electrodes¶
This notebook demonstrates calculating anatomical information about intracranial electrodes and plotting on the Freesurfer average brain.
# Author: Gavin Mischler
#
# License: MIT
import numpy as np
import matplotlib.pyplot as plt
import naplib as nl
from naplib.localization import Brain
from naplib.visualization import plot_brain_elecs, plot_brain_overlay
Load freesurfer fsaverage data if we don't have it
import os
import mne
os.makedirs('./fsaverage', exist_ok=True)
mne.datasets.fetch_fsaverage('./fsaverage/')
Attempting to create new mne-python configuration file:
/home/docs/.mne/mne-python.json
Could not read the /home/docs/.mne/mne-python.json json file during the writing. Assuming it is empty. Got: Expecting value: line 1 column 1 (char 0)
179 files missing from root.txt in /home/docs/checkouts/readthedocs.org/user_builds/naplib-python/checkouts/latest/examples/brain_plotting/fsaverage
Downloading missing files remotely
Downloading data from 'https://osf.io/download/3bxqt?version=2' to file '/tmp/tmp9xet74j2/temp.zip'.
0%| | 0.00/196M [00:00<?, ?B/s]
0%| | 179k/196M [00:00<01:52, 1.74MB/s]
1%|▍ | 2.08M/196M [00:00<00:16, 11.8MB/s]
5%|█▉ | 10.1M/196M [00:00<00:04, 43.0MB/s]
10%|███▋ | 19.0M/196M [00:00<00:02, 60.8MB/s]
14%|█████▏ | 26.8M/196M [00:00<00:02, 67.0MB/s]
18%|██████▊ | 35.3M/196M [00:00<00:02, 73.2MB/s]
22%|████████▎ | 42.7M/196M [00:00<00:02, 73.2MB/s]
26%|█████████▋ | 50.0M/196M [00:00<00:01, 73.1MB/s]
30%|███████████▍ | 58.8M/196M [00:00<00:01, 77.9MB/s]
34%|████████████▉ | 66.9M/196M [00:01<00:01, 78.7MB/s]
39%|██████████████▋ | 75.6M/196M [00:01<00:01, 80.9MB/s]
43%|████████████████▎ | 83.7M/196M [00:01<00:01, 81.0MB/s]
47%|█████████████████▉ | 92.5M/196M [00:01<00:01, 83.1MB/s]
52%|████████████████████ | 101M/196M [00:01<00:01, 82.3MB/s]
56%|█████████████████████▋ | 109M/196M [00:01<00:01, 80.3MB/s]
60%|███████████████████████▎ | 117M/196M [00:01<00:00, 80.6MB/s]
64%|████████████████████████▉ | 125M/196M [00:01<00:00, 76.2MB/s]
68%|██████████████████████████▍ | 133M/196M [00:01<00:00, 76.0MB/s]
72%|████████████████████████████ | 141M/196M [00:01<00:00, 77.7MB/s]
76%|█████████████████████████████▋ | 149M/196M [00:02<00:00, 77.2MB/s]
80%|███████████████████████████████▍ | 157M/196M [00:02<00:00, 79.6MB/s]
85%|█████████████████████████████████ | 166M/196M [00:02<00:00, 81.6MB/s]
89%|██████████████████████████████████▊ | 175M/196M [00:02<00:00, 83.1MB/s]
94%|████████████████████████████████████▌ | 183M/196M [00:02<00:00, 84.0MB/s]
98%|██████████████████████████████████████▎| 192M/196M [00:02<00:00, 84.8MB/s]
0%| | 0.00/196M [00:00<?, ?B/s]
100%|████████████████████████████████████████| 196M/196M [00:00<00:00, 827GB/s]
Extracting missing files
Successfully extracted 179 files
10 files missing from bem.txt in /home/docs/checkouts/readthedocs.org/user_builds/naplib-python/checkouts/latest/examples/brain_plotting/fsaverage/fsaverage
Downloading missing files remotely
Downloading data from 'https://osf.io/download/7ve8g?version=4' to file '/tmp/tmp8f7ktcyo/temp.zip'.
0%| | 0.00/239M [00:00<?, ?B/s]
0%| | 170k/239M [00:00<02:20, 1.70MB/s]
1%|▎ | 2.01M/239M [00:00<00:20, 11.5MB/s]
4%|█▋ | 10.5M/239M [00:00<00:05, 45.1MB/s]
8%|███ | 19.4M/239M [00:00<00:03, 62.5MB/s]
12%|████▌ | 28.4M/239M [00:00<00:02, 72.2MB/s]
16%|█████▉ | 37.3M/239M [00:00<00:02, 78.1MB/s]
19%|███████▎ | 46.3M/239M [00:00<00:02, 82.0MB/s]
23%|████████▊ | 55.2M/239M [00:00<00:02, 84.0MB/s]
27%|██████████▏ | 64.1M/239M [00:00<00:02, 85.7MB/s]
31%|███████████▋ | 73.1M/239M [00:01<00:01, 87.0MB/s]
34%|█████████████ | 82.1M/239M [00:01<00:01, 87.8MB/s]
38%|██████████████▍ | 91.1M/239M [00:01<00:01, 88.4MB/s]
42%|████████████████▎ | 100M/239M [00:01<00:01, 88.8MB/s]
46%|█████████████████▊ | 109M/239M [00:01<00:01, 89.1MB/s]
49%|███████████████████▎ | 118M/239M [00:01<00:01, 89.2MB/s]
53%|████████████████████▋ | 127M/239M [00:01<00:01, 89.4MB/s]
57%|██████████████████████▏ | 136M/239M [00:01<00:01, 89.5MB/s]
61%|███████████████████████▋ | 145M/239M [00:01<00:01, 89.3MB/s]
64%|█████████████████████████ | 154M/239M [00:01<00:00, 89.0MB/s]
68%|██████████████████████████▌ | 163M/239M [00:02<00:00, 88.9MB/s]
72%|████████████████████████████ | 172M/239M [00:02<00:00, 88.1MB/s]
76%|█████████████████████████████▍ | 180M/239M [00:02<00:00, 88.3MB/s]
79%|██████████████████████████████▉ | 189M/239M [00:02<00:00, 88.5MB/s]
83%|████████████████████████████████▎ | 198M/239M [00:02<00:00, 86.4MB/s]
87%|█████████████████████████████████▊ | 207M/239M [00:02<00:00, 87.1MB/s]
90%|███████████████████████████████████▎ | 216M/239M [00:02<00:00, 87.6MB/s]
94%|████████████████████████████████████▋ | 225M/239M [00:02<00:00, 83.3MB/s]
98%|██████████████████████████████████████▏| 234M/239M [00:02<00:00, 85.1MB/s]
0%| | 0.00/239M [00:00<?, ?B/s]
100%|████████████████████████████████████████| 239M/239M [00:00<00:00, 941GB/s]
Extracting missing files
Successfully extracted 10 files
PosixPath('/home/docs/checkouts/readthedocs.org/user_builds/naplib-python/checkouts/latest/examples/brain_plotting/fsaverage/fsaverage')
Create a brain with the pial surface for computing metrics
brain = Brain('pial', subject_dir='./fsaverage/').split_hg('midpoint').split_stg().simplify_labels()
# Specify the coordinates of 30 electrodes in fsaverage space
coords = np.array([[-47.281147 , 17.026093 , -21.833099 ],
[-48.273964 , 16.155487 , -20.162935 ],
[-51.101261 , 13.711058 , -16.258459 ],
[-55.660889 , 9.761111 , -12.340655 ],
[-58.733326 , 6.046287 , -9.626602 ],
[-60.749279 , 2.233287 , -8.044459 ],
[-61.26712 , -1.939675 , -8.582445 ],
[-63.686226 , -10.447982 , -0.445693 ],
[-63.453224 , -9.826311 , 1.095302 ],
[-48.792809 , 15.73144 , -19.34193 ],
[-51.336754 , 13.27527 , -15.57861 ],
[-53.301971 , 11.016301 , -12.48259 ],
[-55.044659 , 9.894337 , -11.228349 ],
[-57.597462 , 6.753941 , -8.082416 ],
[-60.594891 , 2.579503 , -6.884331 ],
[-63.078999 , -8.770401 , -1.878142 ],
[-67.419235 , -26.153931 , -1.260003 ],
[-60.28742599, -11.71243477, 5.62593937],
[-63.12403107, -12.37896156, 4.09772062],
[ 64.44213867, -3.16063929, -6.95104313],
[ 61.58537674, -23.53317833, -3.20349312],
[ 69.31034851, -18.18317795, 1.97798777],
[ 69.0439682 , -18.64465904, 1.2625511 ],
[ 68.32962799, -20.90372849, -0.25190961],
[ 59.79437256, -23.76178932, -3.52095652],
[-56.57900238, -9.23060513, -7.33194447],
[-58.861763 , -11.2859602 , -6.18047237],
[-61.13874054, -11.35863781, -4.49999475],
[-60.82435989, -8.91696072, -3.20156574],
[-61.00576019, -7.45676041, -3.06485367]])
isleft = coords[:,0]<0
Get anatomical labels for each electrode
['mSTG' 'mSTG' 'mSTG' 'mSTG' 'mSTG' 'mSTG' 'mSTG' 'mSTG' 'mSTG' 'mSTG'
'mSTG' 'mSTG' 'mSTG' 'mSTG' 'mSTG' 'mSTG' 'pSTG' 'mSTG' 'mSTG' 'mSTG'
'pSTG' 'pSTG' 'pSTG' 'pSTG' 'pSTG' 'mSTG' 'mSTG' 'mSTG' 'mSTG' 'mSTG']
Compute the distance of each electrode from posteromedial HG along the cortical surface
dist_from_HG = brain.distance_from_region(coords, isleft, region='pmHG', metric='surf')
print(dist_from_HG)
[52.67211969 50.86446306 46.6258215 42.63758415 39.27460724 37.07385395
36.85639736 27.13146072 25.69458138 49.98035531 45.88171027 42.72107443
41.56546499 37.83171644 35.98377774 28.77996954 32.920761 19.91693572
22.38750331 36.74425306 41.73006064 30.15477367 30.96079365 33.36495598
43.59780259 37.00318716 34.83448731 32.22574902 31.71335103 30.79269113]
Smoothly interpolate values over the brain's surface from electrodes
# As an example, we will use the y coordinate of the electrode
values_per_electrode = coords[:,1] - coords[:,1].min()
# Interpolate onto only the temporal lobe, using the 5 nearest neighbor interpolation with
# a maximum distance of 10mm
brain.interpolate_electrodes_onto_brain(coords, values_per_electrode, isleft, roi='temporal', k=5, max_dist=10)
# Plot the overlay for just the left hemisphere
fig, axes = plot_brain_overlay(brain, cmap='Reds', view='lateral', figsize=(12,6), hemi='lh')
plt.show()
Create a brain with the inflated surface for plotting
Plot electrode locations with matplotlib, and color the electrodes by their distance from HG
fig, axes = plot_brain_elecs(brain, coords, isleft, values=dist_from_HG, hemi='lh', view='lateral')
plt.show()
Plot electrodes with an interactive plotly figure, and instead of coloring them by a value, we will color them by a custom color for each electrode. Some common use-cases for this might be to color electrodes based on the identity of the subject they came from when pooling electrodes, or by a categorical variable. In this case, we color them by the anatomical labels they were assigned to (black for posterior STG and red for middle STG)
colors = ['k' if lab == 'pSTG' else 'r' for lab in anatomical_labels]
fig, axes = plot_brain_elecs(brain, coords, isleft, colors=colors, backend='plotly')
fig.write_html("interactive_brain_plot.html") # save as an interactive html plot
fig.show() # show the interactive plot in the notebook
Color certain brain regions by their label
brain.paint_overlay('mSTG', -3)
brain.paint_overlay('pSTG', 3)
brain.paint_overlay('MTG', 1)
fig, axes = plot_brain_overlay(brain, view='lateral')
plt.show()
Plot default Destrieux Atlas region labels overlaid on the brain
brain = Brain('pial', subject_dir='./fsaverage/')
brain.lh.overlay = brain.lh.labels
brain.rh.overlay = brain.rh.labels
fig, axes = plot_brain_overlay(brain, cmap='tab20')
plt.show()
Load and plot Glasser Atlas region labels overlaid on the brain
brain = Brain('pial', atlas='HCPMMP1', subject_dir='./fsaverage/')
brain.lh.overlay = brain.lh.labels
brain.rh.overlay = brain.rh.labels
fig, axes = plot_brain_overlay(brain, cmap='prism')
plt.show()
Total running time of the script: (3 minutes 22.522 seconds)