Package: neural

Methods for neural data processing.

This module contains algorithms that are specific to neural time series processing, mainly pertaining to electrophysiology (EXG). The main use cases here are state-of-the-art artifact removal as well as domain-specific feature extraction, for example the powerful common spatial pattern (CSP) algorithm in various forms. Most of these nodes act on time series, which are expected to have a time axis and a space axis (e.g., channel). Some nodes can also process segmented data, which is expected to additionally have an instance axis (e.g., trials).

Nodes in this package:

• Add or Remove Evoked Activity
  Remove or (re-)add average evoked activity to a given continuous signal.

• Detect Artifacts
  Detect artifacts in a sliding window fashion and mark them up with NaNs.

• Artifact Removal
  Remove various kinds of high-amplitude artifacts from the signal.

• Bad Channel Removal
  Remove channels with abnormal data from a continuous EEG signal.

• Channel Repair
  Repair channels during periods where they record garbage data.

• Classify Independent Components
  Classify independent components using a component classifier.

• Common Spatial Patterns
  Extract signal components whose variance optimally discriminates between two conditions.

• Event Related Potential (Meta)
  A meta-node that segments and calculates the ERPs for a predefined set of event markers.

• Filter Bank Common Spatial Patterns
  Extract signal components across multiple bands whose variance optimally discriminates between two conditions.

• Filter Bank Source Power Comodulation
  Extract signal components across multiple bands whose variance optimally correlates with some target variable.

• Fix Signal Unit
  Infer and then correct the unit of an electrophysiological signal (e.g

• Hypnogram (HMM)
  Calculate a hypnogram (sleep scoring) from EEG observations and a specified initial observation model using a hidden Markov model (HMM) approach.

• Impedance Channel Rejection (Cognionics)
  Remove bad channels based on their impedance.

• Interpolate Missing Channels
  Interpolate missing channels in the given data.

• Phase Amplitude Coupling
  Calculates the degree of dependence of the amplitude signal on the phase of the phase signal.

• Preprocess EEG (Meta)
  Performs a standard EEG preprocessing chain including high/low pass filtering, artifact and bad channel removal, referencing, etc.

• Reject Independent Components
  Automatically reject independent components that have previously been classified into artifact types.

• Remove Bad Time Windows
  This function cuts segments from the data which contain high/low-amplitude artifacts.

• Remove Outlier Trials
  Removes trials (instances) that have abnormally large signal values.

• Replace Thresholded Data
  Replace samples in the signal data with NaN where the value of a provided measure for that same time point and channel has a value above (or below) a specified threshold.

• Source Power Comodulation
  Extract signal components whose variance is is maximally correlated to a target variable of interest.

• Spatio-Spectral Decomposition
  Extract components that are most representative of a given "peak" signal.

• Spectrally Weighted Common Spatial Patterns
  Extract spatio-spectral signal components whose variance optimally discriminates between two conditions.