brain-states

This is a toolbox for computing EEG brain states by using wSMI, sliding window and k-means clustering

Required

You will need to have the following packages and programs

• Python 3.x
• NumPy (https://numpy.org/)
• SciPy (https://www.scipy.org/)
• h5py (https://www.h5py.org/)
• MNE (https://mne.tools/stable/index.html)
• NICE Tools (https://github.com/nice-tools/nice)
• MATLAB
• EEGLAB Toolbox (https://sccn.ucsd.edu/eeglab/index.php)

Functions

• filtering: Applies a bandpass filter and subsampling
• stationary-wsmi: Computes the wSMI matrix for the whole session
• dynamic-wsmi: Uses the sliding window technique and computes a wSMI matrix for every window
• k-means: Using the result from dynamic-wsmi the windows are classified into centroids using the k-means clustering algorithm. It also computes the probability distribution of the states and transition probability between states at t and t+1

Examples

You can find examples for every function in the "example" folder.