Information Technology Reference
In-Depth Information
properly setup the EEG cap. The approach presented in this chapter applies a
preprocessing method for EEG signals based on the use of discrete wavelet
decomposition (DWT) to extract the energy of each frequency in the signal.
Then, a linear regression is performed on the energies of some of these
frequencies and the slope of this regression is retained. A genetic algorithm (GA)
is used to optimize the selection of frequencies on which the regression is
performed and to select the best recording electrode. Results show that the
proposed strategy derives accurate predictive models of alertness.
9.1
Introduction
Over the last decade, human
computer interaction (HCI) has grown and matured as
a field. Gone are the days when only a mouse and keyboard could be used to
interact with a computer. The most ambitious of such interfaces are brain
-
computer
interaction (BCI) systems. The goal in BCI is to allow a person to interact with an
arti
-
cial system using only brain activity. The most common approach toward BCI
is to analyze, categorize, and interpret electroencephalographic (EEG) signals, in
such a way that they alter the state of a computer.
In particular, the objective of the present work is to study the development of
computer systems for the automatic analysis and classi
cation of mental states of
vigilance; i.e., a person
is state of alertness. Such a task is relevant to diverse
domains, where a person is expected or required to be in a particular state. For
instance, pilots, security personnel, or medical staffs are expected to be in a highly
alert state, and a BCI could help con
'
rm this or detect possible problems.
It is possible to assume that the speci
c topic presented in this chapter lies
outside the scope of this topic, entitled
“
Guide to Brain-Computer Music Inter-
facing.
Nevertheless, from our point of view, many tasks have to be accomplished
before any interaction between a person
”
s brain and music can be done by using
EEG signals. Suppose that we wish to develop a musical instrument that can
generate music that is speci
'
cally related to the alertness of a subject. For such a
system, a
first objective should be to classify the EEG signals of a subject based on
different levels of alertness. In order to reach this objective, informative features
have to be extracted, particularly since processing raw EEG data is highly
impractical, and then proceed to a
cation step using relevant mathe-
matical concepts. However, this problem is by no means a trivial one. In fact, EEG
signals are known to be highly noisy, irregular, and tend to vary signi
final classi
cantly from
person to person, making the development of general techniques a very dif
cult
scienti
find a method that is adaptable to
different persons and that it provides a rapid and accurate prediction of the alertness
state. For instance, a similar problem is presented by Lin et al. (
2010
), the authors
developed a feature extraction and classi
c endeavor. Then, it is important to
cation approach to classify emotional
states and build an immersive multimedia system, where a user
'
s mental states
in
uences the musical playback. Examples such as these illustrate the importance of
