Affordable brain-computer interfaces (BCIs) working outside the laboratory environment
are still rare. Their most limiting factors include low classification
accuracy, information transfer bit rate, low variability of used approaches, and
closeness of the hardware and software components of the system. The presented
BASIL project has focused on designing, developing, and testing an affordable
BCI system built on low-cost hardware and open-source software components.
It provides people with motor impairments with an opportunity to control their
basic home environment.
The concept of the BASIL prototype follows the best practices that are
known within the construction of BCI systems, adds the concept of the cloud
for remote BCI computations, relies on testing and customization of the whole
system to the needs of individuals, and focuses on the solution affordable for
ordinary users. The core components of the BASIL project solution include
hardware components for signal acquisition and software components for local
execution of online BCIs.
The BASIL system was tested on ten participants in laboratory conditions
using various BCI paradigms. We failed to evoke a reliable P300 component
with eight-trial averages. Eyes blinks, alpha activity, and steady-state visually
evoked potentials were clearly observable. Dry electrodes with long pins were
preferred by most users. Out of ten participants, six could control the system
online, achieving more than 70 % accuracy.
The results show that a successful BCI system can be built on low-cost hardware
for EEG signal acquisition and amplification. The benefits and weaknesses
of known BCI paradigms for such a system have been identified. The current
solution, the affordable BASIL BCI system prototype, is prepared for further
community development and testing.