Note
Click here to download the full example code
In this example, we load in a single subject example, remove electrodes that exceed a kurtosis threshold (in place), load a model, and predict activity at all model locations. We then convert the reconstruction to a nifti and plot 3 consecutive timepoints first with the plot_glass_brain and then create .png files and compile as a gif.
# Code source: Lucy Owen & Andrew Heusser
# License: MIT
# load
import supereeg as se
# load example data
bo = se.load('example_data')
# load example model
model = se.load('example_model')
# the default will replace the electrode location with the nearest voxel and reconstruct at all other locations
reconstructed_bo = model.predict(bo)
# print out info on new brain object
reconstructed_bo.info()
# convert to nifti
reconstructed_nifti = reconstructed_bo.to_nii(template='gray', vox_size=20)
# make gif, default time window is 0 to 10, but you can specifiy by setting a range with index
# reconstructed_nifti.make_gif('/your/path/to/gif/', index=np.arange(100), name='sample_gif')
Total running time of the script: ( 0 minutes 0.000 seconds)