Revision history [back]

In order to create an ERDS (Event-Related Desynchronization/Synchronization) graph using numpy or rawarray files, you first need to have the data in the form of EEG (Electroencephalography) signals. There are several steps involved in creating an ERDS graph:

1. Preprocessing the EEG data: This involves removing any noise and artifacts from the data, and filtering it to remove any unwanted frequencies.

2. Segmenting the data into epochs: The EEG data is divided into epochs or time windows, typically around 1-3 seconds long, aligned with the time of the event being studied. These epochs are typically baseline-corrected to remove any DC shift.

3. Computing the power spectral density: For each epoch, the power spectral density is calculated for each frequency band of interest (such as alpha, beta, gamma).

4. Computing the ERD/ERS: The Event-Related Desynchronization/Synchronization is calculated by comparing the power spectral density during the event with the baseline period immediately before the event. ERD/ERS is usually expressed as a percentage change in power.

5. Plotting the ERDS graph: The ERDS values can be plotted as a color-coded heatmap or contour plot, with time on the x-axis, frequency on the y-axis, and ERDS values represented as color.

To perform these steps using numpy or rawarray files, you can use Python libraries such as MNE (Martinos Center for Biomedical Imaging, Massachusetts General Hospital) or EEGPy (EEG Analysis in Python). These libraries have functions for loading, preprocessing, and segmenting EEG data, as well as computing the power spectral density and ERD/ERS values. They also have built-in functions for plotting ERDS graphs.