Symbolic Transfer Entropy

Symbolic Transfer Entropy Parameters :

1. Windows Size: Windows size used to calculate each segment for the symbolic transfer entropy analysis.
2. Number of Windows: Number of windows of the given size inputted above. Make sure the numbers of windows times the windows size is not greater than the length of the EEG.
3. Source Channels: The channels from which a signal could originate.
4. Sink Channels: The channels that receive could receive a signal from the source channels.
5. Dim: Dim represent the embedding dimension2.
6. Tau: Tau represent the time delay2. Both variables are used to convert the transfer entropy into its symbolic form.
7. Print check box: If this is selected when the pipeline will be done computing the symbolic transfer entropy it will output it on the screen in the command windows. Because the output of this analysis technique is a structure file the output might get messy and span a great length of the command windows.
8. Save check box: When done, the pipeline will save the Symbolic Transfer Entropy data as a structure file in two form (fig. 24). The first form is a long storage form looking like this BandpassDay-month-year_time_ste.mat and another one that is short term and that will be overwritten each time a STE analysis with a same bandpass will be done. It will look like this: Bandpassste.mat. Also, a text file containing the input will be saved using the same name as the structure.

Bandpass Menu :

In the bandpass menu you can select at which frequency you want to filter your EEG data in order to do the Symbolic Transfer Entropy analysis. If you select more than one frequency pass, the analysis will be repeated for each bandpass,

• Full: Full band goes from 1Hz to 50Hz.
• Delta: Delta band goes from 1Hz to 4Hz.
• Theta: Theta band goes from 4Hz to 8Hz.
• Alpha: Alpha band goes from 8Hz to 13Hz.
• Beta: Beta band goes from 13Hz to 30Hz.
• Gamma: Gamma band goes from 30Hz to 50Hz.

Symbolic Transfer Entropy Examples

Up in a few...