Dry andWater-Based EEG Electrodes in SSVEP-Based BCI Applications - Biomedical Engineering Systems and Technologies - page 6

(a) (b) (c)

Fig. 1. EEG setups: a) Dry contact electrodes with pins integrated in the EEG headband; b) A
water-based electrode and its components; c) Inner and outer side of the head cap with electrodes
that need to be filled with conductive gel

2 mm long rigid pins. These electrodes are attached to a soft textile patch which is con-
nected to the elastic head band using six Velcro straps, as depicted in Figure 1a. Velcro
straps are used to accommodate for head sizes and head shapes of different participants.
The electrodes are connected using shielded cables to the TMSi Porti acquisition sys-
tem. The setup is mounted on a participant's head similarly as a normal headband. To
the best of our knowledge this is the first 'easy-to-mount' EEG setup with multitude of
dry contact electrodes aimed at measuring SSVEP response.
For water-based setup we used electrodes that require tap water instead of electrolytic
gel. First tests with such electrodeas are reported in [41]. Electrodes are made from a
silver-chloride pallet and rolled up cotton, as shown in Figure 1b, and are connected to
the Porti system via shielded cables. Commercially available EEG head cap with the
screwing mechanism was used to position the water-based sensors on the head.
The setup for measuring brain signals using conductive gel was prepared using a
standard 32 channel head cap (depicted in Figure 1c) for the usage with the TMSi Porti
EEG acquisition system. Shielded cables were also used, as in the case of dry and water-
based electrodes. Out of the 32 channels, only the 8 selected channels were filled with
conductive gel (Signa gel from Parker Laboratories). In the preparation step we did not
used any skin abrasion or clean up procedure in order to reproduce as closely as possible
the setup in daily life applications.

3.3 Methods

This section describes the design of the experimental evaluation, the algorithms we used
for handling artifacts, the estimation of the magnitude of the SSVEP response, and the
manner in which we estimated the performance and comfort of different setups.

Study Design. Six participants (4 males and 2 females) aged 24, 26, 28, 29, 31, and
32, were recruited for the experiment. Participants were selected to cover different hair
characteristics ranging from short, sparse, and thin hair, to long, dense, and thick hair.
The participants were informed about the experiment and they all signed an informed
consent before the start of the study. Special emphasis was put on verifying that the
participants did not have any history of epileptogenic episodes or discomforts due to