Skip to content

This function outputs a set of spectrally important channels based on a predetermined threshold.

License

Notifications You must be signed in to change notification settings

jameswilsenach/SEEG_Channel_Selection

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

5 Commits
 
 
 
 
 
 

Repository files navigation

Spectral Principal Component Analysis for SEEG Channel Selection

Often S/EEG analysis requires the extraction of a small set of important channels for further analysis. This method uses Spectral Principal Component Analysis (SPCA), also called Spectral Proper Orthogonal Decomposition (SPOD) to determine which channels are the most important based on a predetermined energy cutoff.

Installation

This function requires MATLAB v2019a or later as well as the implementations of SPCA, spod.m and inverse SPCA, invspod.m by Oliver Schmidt, saved to your MATLAB path, which is available here: https://www.mathworks.com/matlabcentral/fileexchange/65683-spectral-proper-orthogonal-decomposition-spod

Instructions

In order to run the function you will require an appropriate SEEG file saved as a MATLAB data array with dimensions RxMxT, where R is the number of repeated trials, M is the number of S/EEG channels and T is the length of each trial (recommended to be at least 20s or greater). The current model assumes an energy cutoff of 75% but this can be changed in options. Increasing the energy threshold will decrease the selectivity of the method.

About

This function outputs a set of spectrally important channels based on a predetermined threshold.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages