Sensor Position Dependent Factors#
Having extracted patient’s anatomy in section Anatomy Extraction, the following steps compute the co-variability between individual M/EEG channels and the co-registration between M/EEG sensor position and anatomy. Unlike anatomy, these may have to be recomputed for a number of scenarios, including refreshing of EEG gel, EEG electrode cap replacements, patient change, moving significantly within the MEG, changing body position in the MEG.
Sensor noise covariance#
The sensor noise covariance describes how strongly a signal change in one channel is reflected in another channel (assuming a linear relationship). This relationship is critical to ultimately identify those channels that contribute to a change of the observed signals.
The following code-block outlines how to compute the sensor noise covariance.
import finnpy.src_rec.sen_cov
finnpy.src_rec.sen_cov.run(sensor_data, fs, signal_type,
valid_channels, ch_names, ch_types,
method = None, float_sz = 64, epoch_sz_s = 0.2, method_params = None,
fast_eigendecomp_path = "../FinnPy_speedups/Release/FinnPy_speedups.so"):
M/EEG co-registration#
The M/EEG co-registration describes the spatial transformation between M/EEG sensors and the patient’s anatomy.
The following code-block outlines how to compute the M/EEG co-registration.
coreg = finnpy.src_rec.coreg.run(subj_name, anatomy_path, signal_type, use_nasion = True, rec_info = None)
Next steps#
This concludes the sensor position focused section. The inverse model is calculated in the next section Downstream Processing.