Information Technology Reference
In-Depth Information
Fig. 4.3: The second selected EEG segment and ICA results from patient 1. (
a
)
The 15-s time window (152-167 s) used to display results in (
b
) and (
d
). (
b
)The
illustration of a 15-s segment where signals in the
shaded areas
were severely con-
taminated by large eye movements and environmental noises. (
c
) The topographical
maps generated at four peak time points p1, p2, p3, and p4 (
vertical lines
in
b
)of
four waves in IC4 at 154, 156.6, 159.9, and 162.9 s. (
d
) The 17 decomposed ICs
show that diseased-related pattern was focal triphasic waves (IC4) and the artifacts
were eyeblinks (IC2), eye movements (IC8), and noise (IC15). (
e
) The correspond-
ing spatial maps of IC2, IC4, IC8, and IC15.
To recover the CJD-related patterns from EEG data, we employed the indepen-
dent component analysis (ICA) [11, 23] in this study. ICA has been successfully
applied to remove nonphysiological artifacts from EEG data [14, 15], to segregate
Rolandic beta rhythm from magnetoencephalographic (MEG) measurements of the
right index finger lifting [18], to extract the task-related features from the motor
imagery EEG and the flash visual evoked EEG in the studies of the brain com-
puter interface [10, 17], to analyze the interactions during temporal lobe seizures in
stereotactic depth EEG [22], to separate generalized spike-and-wave discharges into
the primary and secondary bilateral synchrony [13], and to segment spatiotemporal
hemodynamics from perfusion magnetic resonance brain images [16].