This document discusses the challenges of analyzing EEG data collected during long-term ambulatory monitoring, where many types of movement artifacts are present. It describes common artifacts like eye blinks, head movements, chewing, and electrode tapping that can interfere with seizure detection. The document proposes using time-frequency decomposition and statistical feature extraction to differentiate seizures, artifacts, and normal EEG patterns for more accurate epilepsy diagnosis and prediction. Evaluation on simulated seizure and artifact data shows the potential of this artifact removal method to improve seizure detection performance.

1. Md Kafiul Islam
Translational System and Signal Processing Group
Dept. of Electrical & Computer Engineering
National University of Singapore

2.  Purpose:
◦ Diagnosis/Detection of Epilepsy Seizure by Long-term EEG Monitoring (up
to 72 hours)
◦ Early warning of seizures (prediction) onset in order to stop seizure
 Special Features
◦ Continuous monitoring
◦ Wireless EEG (up to 10 m or so)
◦ Usually Dry electrode
◦ Few no. of channels
 typically 4-8 channels
◦ Patient can lead normal daily life
 Except few tasks (e.g. taking bath)

3.  Head Movement ( < 10 Hz )
◦ Shake
◦ Nod
◦ Roll
 Hand Movement
 Eye Movement
◦ Eyebrow Movement
◦ Eye blinking
◦ Horizontal and Vertical Eye Movement
◦ Extra-ocular muscle activity
 Jaw Clenching/Eating/Drinking
 Walking
 Running
 Sleeping
◦ Drowsiness, Light Sleep, Deep Sleep, REM
 Talking
 Working on PC
 Watching TV
 And Many More!
Consequence:
Too Many Motion Artifacts!

Challenge: Differentiate
between artifacts, epileptic
seizures and normal EEG activities!
How …!???

4. Horizontal Eye
Movement (T1-T2)
and
Blink Artifacts
(Frontal channels)
EOG 
*Chang, B. S., Schachter, S. C., & Schomer, D. L. (2005). Atlas of ambulatory EEG.
Amsterdam ; London: Amsterdam ; London : Elsevier Academic Press, c2005.

5. Rhythmic eye flutter can occur at quite high frequencies, in this case 5–8 Hz, and
trigger automated seizure detection algorithms.
This artifact can be distinguished from ictal events in different ways, including the
lack of evolution in frequency or amplitude over time.
Eye Flutter Artifact
recorded by Seizure
Detection Algorithm

6. Electrode Tapping
Artifact
With ambulatory EEG recording, there may be periods in which the patient
is scratching or pressing on the scalp electrodes.

7. Jaw-Clenching
Artifact
 Periods of jaw clenching and biting during ambulatory EEG recording can be associated with artifact.
 This activity is often most prominent over the temporal regions, presumably due to temporalis or
masseter muscle contraction.

8. Chewing Artifact
 Chewing artifact can often have rhythmic features and at times may resemble ictal activity.
 This artifacts are characterized by the rhythmic appearance of muscle artifact centered mostly
over the temporal regions bilaterally.

9. Chewing Artifact
Recorded by
Spike Detection
Algorithm
 Chewing artifact can have rhythmic and sharp features, and can thus be picked up by automated
detection algorithms.
Here, the spike detections at 23:53:23 and at 23:53:34 capture high frequency sharp waveforms
that are artifactual in nature, in association with a nighttime snack.

10. Dry Electrode Artifact
 One disadvantage of ambulatory EEG monitoring is the inability of technologists to check on
recording quality frequently in real time and to respond with appropriate technical adjustments as
needed throughout the day, as would typically occur in an inpatient monitoring unit.
 Here, a dry electrode at F8 causes an artifact seen throughout this excerpt.

11. Forehead Rubbing
Artifact
 As with many other patient movement artifacts, rhythmic rubbing artifact can be a potentially
confusing finding on ambulatory EEG monitoring, particularly in the absence of concomitant video
recording. Here, rhythmic rubbing of the forehead produces an artifact in the anterior channels,
extending from the beginning of this excerpt through 20:50:16.
 Differentiation of such artifact from an ictal event, based on the lack of evolution of the artifact
in frequency or amplitude, is critical.

12. Pulse Artifact
 Pulse artifact is caused by the rhythmic movement of scalp electrodes, usually over the
temporal regions, by the pulsations of blood through vessels close to the skin, such as the
superficial temporal artery.
 It is to be distinguished from the more common electrocardiographic (EKG) artifact in that pulse
artifact is typically not as sharply contoured, resembles a slow wave, and follows the QRS complex
rather than being simultaneous to it.
 Here, pulse artifact is seen in channels 9 and 10 over the left temporal region throughout this
entire excerpt.

13.  Delta: (< 4 Hz)
 Theta: (4 - 8 Hz)
 Alpha: (8 - 13 Hz)
 Beta: (14 - 30 Hz)
 Gamma: (> 30 Hz)
 Mu: (7.5 - 12.5 Hz)
Gamma

14.  Use of an extra sensor
◦ Adaptive filtering
 Accelerometer
 Gyroscope
 Contact Impedance Measurement
 A-priori user input required
◦ Wiener/Kalman/Particle Filtering
 Blind Source Separation
◦ ICA/CCA/MCA
 Time-series Analysis
◦ Wavelet transform / STFT
 Empirical Technique
◦ HHT/EMD/EEMD

15. Artifact Generation Protocol (From Literature)

16.  Domain
◦ Time
◦ Frequency
◦ Wavelet
 Statistical Features
◦ Entropy
 ApEn, MSE, CMSE, MMSE, etc.
◦ Kurtosis
◦ Skew
◦ RMS/Variance/S.D.
◦ Max/Min/Mean
◦ Maxima/Minima
High Frequency Oscillations
(HFO)
80 Hz – 200 Hz
Ripple
200 Hz – 600 Hz
Typical Scalp EEG!
0.05Hz – 128 Hz
Seizure Appears in Scalp EEG
0.5 Hz - 29 Hz [1,2]
1. J. Gotman, “Automatic recognition of epileptic seizures in the EEG,”
Electoencephalogr Clin. Neurophysiol., vol. 54, pp. 530–540, 1982
2. M. E. Saab and J. Gotman, “A system to detect the onset of epileptic
seizures in scalp EEG,” Clin. Neurophysiol., vol. 116, pp. 427–442,
2005.

17. Seizure Appears between 0.5 Hz to
29 Hz
Raw EEG
D1
(64 – 128 Hz)
A1
(0 – 64 Hz)
D2
(32 – 64 Hz)
A2
(0 – 32 Hz)
D3
(16 – 32 Hz)
A3
(0 – 16 Hz)
D4
(8 – 16 Hz)
A4
(0 – 8 Hz)
D5
(4 –8 Hz)
A5
(0 – 4 Hz)
D6
(2 – 4 Hz)
A6
(0 – 2 Hz)
D7
(1 – 2 Hz)
A7
(0 – 1 Hz)
D8
(0.5 – 1 Hz)
A8
(0 – 0.5 Hz)
Multilevel Wavelet Decomposition
Fs = 256 Hz, L = 8, Haar Wavelet

19. Artifact Removal Applied on Simulated Data
o Simulated Clean EEG
(downloaded from [1])
o Simulated Seizure
(downloaded from [2])
o Artifact Simulation
(to mimic extreme ambulatory condition)
[1]. http://www.cs.bris.ac.uk/~rafal/phasereset/
[2]. Newborn EEG Seizure Simulation

20. Performance Evaluation of Proposed Artifact Removal
-0.8 -0.7 -0.6 -0.5
20
25
30
35
40
45
50
55
Artifact SNR in dB
%artifactreduction,lamda
o The target is to remove
artifacts as much as possible
o Artifact removal should
enhance the performance of
later signal processing stage
(e.g. increase seizure detection
rate)
o Average amount of artifact
reduction is around 35% when
artifact SNR is around 0 dB
o Not all artifacts affect seizure
detection performance

21. EEG Features before and after Artifact Removal
Features Extracted:
(i) Entropy (ii) Kurtosis (iii) Line Length (iv) Peak
(v) NEO (vi) Variance (vii) FFT (viii) FFT Peak
Note: The features between seizure and non-seizure data are more separable after artifact removal which
suggests that it increases the detection rate and minimizes false alarms (false alarms are due to artifacts).

22.  Characterize movement artifacts by doing EEG
experiments
 Use real seizure data marked by the epilepsy
specialists
 Improve the algorithm by avoiding ICA
 Use classification techniques to differentiate
seizure from non-seizure data and evaluate the
performance of proposed algorithm