Single channel approach for filtering electroencephalographic signals strongly contaminated with facial electromyography

Data: EEG contaminated by facial EMG

The dataset contains the EMG and EEG signals collected for the study of contamination by facial EMG on EEG.

For each volunteer (S1 to S10), there are five “.mat” files, referring to the five muscles activated to generate contamination of facial EMG on EEG.

For each muscle there are some variables:

1. amplifier_data:

It is a matrix with the EMG (card A) and EEG (card B) signals collected. The

amplifier_channels structure and the t_amplifier vector refere to amplifier_data

matrix.

2. board_dig_in_data:

As expected in the experimental protocol, the collection is performed with the

eyes in two states: only open and only closed. The board_dig_in_data variable

contains digital values (0 or 1). The first moment containing 1s marks the

beginning of the eye-open-only state and the second moment indicates the eye-closed-only state. The board_dig_in_channels structure and the t_dig vector refere

to board_dig_in_data variable.

3. supply_voltage_data:

Data collection is performed by two acquisition cards, A(EMG) and B(EEG). The

supply voltage of each card is monitored, and the measured values are stored in

this variable. The supply_voltage_channels structure and the t_supply_voltage

vector refere to supply_voltage_data variable.

EMG and EEG data

The data of interest for the collection are in the amplifier_data variable, a matrix

with n rows and m columns, where n is the number of channels (EMG and EEG) and m is the number of data acquired at the sampling frequency for the entire collection time interval. EMG channels are collected by “Port A”, and EEG channels are collected by “Port B”, but all EMG and EEG data are in the same amplifier_data variable. The name “Port A” or “Port B” comes from the choice of cable used to connect the acquisition cards to the RHD2000 USB interface system. Thus, based on the choice adopted in connecting the EMG and EEG cards in this research, the amplifier_data variable has the first lines belonging to “Port A” (EMG) and the other lines of the matrix being from “Port B” (EEG). The amplifier_channels structure indicates which card was used with the name “Port A” and “Port B” in the port_name field.

EMG data

The muscle signal data does not have a specific and fixed channel between collections. Thus, the channel number for a muscle must be verified at each data collection. The information for each channel must be consulted in the amplifier_channels structure at the custom_channel_name field.

EEG data

A bipolar 16-channel acquisition card acquires EEG signals. The channels of each

EEG sensor are the same for all collections, thanks to the use of the adapted EEG cap, and the connection of the EEG signals card is always carried out on the same cable B. The EEG cap has bipolar transverse mounting configuration.

EEG sensors channel list, channel x EEG sensor:

Ch00: F3-Fz;  Ch01: F4-F8;  Ch02: F7-F3;  Ch03: Fp1-Fp2;

Ch04: T3-C3;  Ch05: C4-T4;  Ch06: T5-P3;  Ch07: P4-T6;

Ch08: Fp2-F8;  Ch09: F7-Fp1;  Ch10: Fz-F4;  Ch11: Cz-C4;

Ch12: O1-O2;  Ch13: P3-Pz;  Ch14: Pz-P4;  Ch15: C3-CZ.