Independent Component Analysis

M/EEG signal decomposition using Independent Component Analysis (ICA). ICA is sensitive to low-frequency drifts and therefore requires the data to be high-pass filtered prior to fitting. Typically, a cutoff frequency of 1 Hz is recommended.

Inputs

Outputs

  • ICA: Object of Independent Component Analysis

Use

widgets/images/ica1.png

  1. Number of components - Number of principal components that are passed to the ICA algorithm during fitting.
  2. Random state - Random state to initialize ICA estimation for representativ results.
  3. Maximum number of iterations - Maximum number of iterations during fit.

Example

widgets/images/exa5work.png

widgets/images/exa5plot1.png

widgets/images/exa5plot2.png

More information here and here.