supereeg.Brain¶

class supereeg.Brain(data=None, locs=None, sessions=None, sample_rate=None, meta=None, date_created=None, label=None, kurtosis=None, kurtosis_threshold=10, minimum_voxel_size=3, maximum_voxel_size=20, filter='kurtosis')[source]¶

Brain data object for the supereeg package

A brain data object contains a single iEEG subject. To create one, at minimum you need data (samples x electrodes), location coordinates in MNI space and the sample rate of the data. Additionally, you can include a session id. If included, all analyses will be performed within session and then aggregated across sessions. You can also include a meta dict, which can contain any other information that might be useful (subject id, recording params, etc).

Parameters:	data : numpy.ndarray or pandas.DataFrame, supereeg.Model, supereeg.Nifti, or Nifti1Image Samples x electrodes array containing the iEEG data. If data is a model, returns correlation matrix. If data is a nifti image (either supereeg.Nifti or Nifti1Image), returns nifti values as samples by electrodes array. locs : numpy.ndarray or pandas.DataFrame Electrode by MNI coordinate (x,y,z) array containing electrode locations session : str, int or numpy.ndarray Samples x 1 array containing session identifiers for each time sample. If str or int, the value will be copied for each time sample. sample_rates : float, int or list Sample rate (Hz) of the data. If different over multiple sessions, this is a list. meta : dict Optional dict containing whatever you want. date created : str Time created (optional) label : list List delineating if location was reconstructed or observed. This is computed in reconstruction.
Returns:	bo : supereeg.Brain Instance of Brain data object.
Attributes:	data : pandas.DataFrame Samples x electrodes dataframe containing the EEG data. locs : pandas.DataFrame Electrode by MNI coordinate (x,y,z) df containing electrode locations. sessions : pandas.Series Samples x 1 array containing session identifiers. If a single value is passed, a single session will be created. sample_rates : list Sample rate of the data. If different over multiple sessions, this is a list. meta : dict Optional dict containing whatever you want. n_elecs : int Number of electrodes dur : float Amount of data in seconds for each session n_sessions : int Number of sessions label : list Label for each session kurtosis : int Kurtosis threshold filter : ‘kurtosis’ or None If ‘kurtosis’, electrodes that exceed the kurtosis threshold will be removed. If None, no thresholding is applied. minimum_voxel_size : positive scalar or 3D numpy array Used to construct Nifti objects; default: 3 (mm) maximum_voxel_size : positive scalar or 3D numpy array Used to construct Nifti objects; default: 20 (mm)

Methods

apply_filter([inplace])	Return a filtered copy
get_data()	Gets data from brain object
get_locs()	Gets locations from brain object
get_slice([sample_inds, loc_inds, inplace])	Indexes brain object data
get_zscore_data()	Gets zscored data from brain object
info()	Print info about the brain object
next()	Return next sample from Brain object (wrapper for self.__next__)
plot_data([filepath, time_min, time_max, ...])	Normalizes and plots data from brain object
plot_locs([pdfpath])	Plots electrode locations from brain object
resample([resample_rate])	Resamples data
save(fname[, compression])	Save method for the brain object
to_nii([filepath, template, vox_size, ...])	Save brain object as a nifti file.

update_filter_inds
update_info

apply_filter(inplace=True)[source]¶

Return a filtered copy

get_data()[source]¶

Gets data from brain object

get_locs()[source]¶

Gets locations from brain object

get_slice(sample_inds=None, loc_inds=None, inplace=False)[source]¶

Indexes brain object data

Parameters:	sample_inds : int or list Times you wish to index loc_inds : int or list Locations you with to index inplace : bool If True, indexes in place.

get_zscore_data()[source]¶

Gets zscored data from brain object

info()[source]¶

Print info about the brain object

Prints the number of electrodes, recording time, number of recording sessions, date created, and any optional meta data.

next()[source]¶

Return next sample from Brain object (wrapper for self.__next__)

plot_data(filepath=None, time_min=None, time_max=None, title=None, electrode=None)[source]¶

Normalizes and plots data from brain object

Parameters:	filepath : str A name for the file. If the file extension (.png) is not specified, it will be appended. time_min : int Minimum value for desired time window time_max : int Maximum value for desired time window title : str Title for plot electrode : int Location in MNI coordinate (x,y,z) by electrode df containing electrode locations

plot_locs(pdfpath=None)[source]¶

Plots electrode locations from brain object

Colors:

• Observed : Blue
• Removed : Cyan
• Reconstructed : Red

Parameters:	pdfpath : str A name for the file. If the file extension (.pdf) is not specified, it will be appended.

resample(resample_rate=None)[source]¶

Resamples data

Parameters:	resample_rate : int or float Desired sample rate

save(fname, compression='blosc')[source]¶

Save method for the brain object

The data will be saved as a ‘bo’ file, which is a dictionary containing the elements of a brain object saved in the hd5 format using deepdish.

Parameters:	fname : str A name for the file. If the file extension (.bo) is not specified, it will be appended. compression : str The kind of compression to use. See the deepdish documentation for options: http://deepdish.readthedocs.io/en/latest/api_io.html#deepdish.io.save

to_nii(filepath=None, template='gray', vox_size=None, sample_rate=None)[source]¶

Save brain object as a nifti file.

Parameters:

filepath : str Path to save the nifti file template : str, Nifti1Image, or None Template is a nifti file with the desired resolution to save the brain object activity If template is None (default) : Uses gray matter masked brain downsampled to brain object voxel size (max 20 mm) If template is str : Checks if nifti file path and uses specified nifti If not a filepath, checks if ‘std’ or ‘gray’ ‘std’: Uses standard brain downsampled to brain object voxel size ‘gray’: Uses gray matter masked brain downsampled to brain object voxel size If template is Nifti1Image : Uses specified Nifti image

Returns:

nifti : supereeg.Nifti A supereeg nifti object