Skip to content

fieldtrip/contrib-realtime_hilbert

Repository files navigation

Real-time neurofeedback application based on Hilbert phase estimation

This package describes the neurofeedback application that was developed by Marco Rotonda for his MSc thesis "Controllo volontario della sincronizzazione di fase intraemisferica nella banda gamma eeg mediante neurofeedback", which translates to "Voluntary control of intrahemispheric phase sychronization in the EEG gamma band using neurofeedback".

The thesis (in Italian) which describes all details and the results is available here.

The signal analysis details are available here.

Neurofeedback, as the word suggest, is feeding back to the subject it's neuronal activity analyzed by it's EEG activity in (almost) real time. In this training we give back to the subject a visual stimulation (vertical bars) when it's gamma phase synchronization between F3-P3 and F4-P4 increased over the baseline (bars up) or decreased (bars down).

Here is the feedback

feedback interface

If you scroll through the script, you can see that this image is a particular point of view of a 3D visualization. If you change the POV you could see all the data from the last 3 minutes.

Procedure

Get the FieldTripBufferDemo.exe from the workshop/bci2000 folder. Start BCI2000 and modifiy the config. The script works originally with 40 channels (see the variable 'lab' in the realtime_hilbert script), the sampling rate used is 1000Hz. Use a blocksize of 1000 samples. Ensure BCI2000 runs for more than 2 minutes in the application tab. Set config and start BCI2000. Start MATLAB and in the command window type ft_realtime_hilbert.

Now it will start the function ft_realtime_baseline that for 2 minutes will record the subject baseline and store the results for ft_realtime_hilbert.

Basically there are 12 channels: 4 for EEG (I used F3-P3-F4-P4), 4 for the EOG, 4 for the EMG (front and neck). In this script, those are selected from the forty originally recorded channels you can find in the variable 'lab'. The algorithm will care to take away both EOG and EMG artifacts.

History

2009 - first implementation was contributed to the FieldTrip website
2026 - moved from the FieldTrip website to its own repository

About

Realtime neurofeedback application based on Hilbert phase estimation

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors

Languages