SlideShare a Scribd company logo
1 of 90
Two Algorithms for Weakly Supervised
Denoising of EEG Data
Tim Oates
Professor
University of Maryland Baltimore County
The Real Brains
3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 2
Sunil Gandhi
Ph.D. student
Will graduate soon!
Agenda
• Introduction To Artifact Removal From EEG data
• ICA And Multi-instance Learning Solution
• Asymmetric Generative Adversarial Network Solution
• Future Work
• Questions
3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 3
• Introduction To Artifact Removal From EEG data
• ICA And Multi-instance Learning Solution
• Asymmetric Generative Adversarial Network Solution
• Future Work
• Questions
3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 4
Electroencephalography (EEG)
• Electroencephalography (EEG) is used to record electrical
activity in the brain
5
Image Source: https://en.wikipedia.org/wiki/Electroencephalography
3/2/2019 Artifact Removal From Electroencephalography (EEG) Data
Electroencephalography (EEG) Data
• Electroencephalography (EEG) is used to record electrical
activity in the brain
– Seizure detection
6
Page, Adam, Chris Sagedy, Emily Smith, Nasrin Attaran, Tim Oates, and Tinoosh Mohsenin. "A flexible multichannel EEG
feature extractor and classifier for seizure detection." IEEE Transactions on Circuits and Systems II: Express Briefs 62, no. 2
(2015): 109-113.
3/2/2019 Artifact Removal From Electroencephalography (EEG) Data
Electroencephalography (EEG) Data
• Electroencephalography (EEG) is used to record electrical
activity in the brain
– Seizure detection
– Determining cognitive load
7
Bashivan, Pouya, Irina Rish, Mohammed Yeasin, and Noel Codella. "Learning representations from EEG with deep recurrent-
convolutional neural networks." International Conference on Learning Representations 2016.
3/2/2019 Artifact Removal From Electroencephalography (EEG) Data
Electroencephalography (EEG) Data
• Electroencephalography (EEG) is used to record electrical
activity in the brain
– Seizure detection
– Determining cognitive load
– Brain Computer interfaces (BCI)
8
Source: https://www.youtube.com/watch?v=rYwgnFeFqmc
3/2/2019 Artifact Removal From Electroencephalography (EEG) Data
Electroencephalography (EEG) Data
• Electroencephalography (EEG) is used to record electrical
activity in the brain
– Seizure detection
– Determining cognitive load
– Brain Computer interfaces (BCI)
– Driver drowsiness estimation
9
Lance, Brent J., W. David Hairston, Greg Apker, Keith W. Whitaker, Geoff Slipher, Randy Mrozek, Scott E. Kerick, Jason Metcalfe,
Christopher Manteuffel, and Matthew Jaswa. 2012 year-end report on neurotechnologies for in-vehicle applications. No. ARL-SR-267.
ARMY RESEARCH LAB ABERDEEN PROVING GROUND MD, 2013.
3/2/2019 Artifact Removal From Electroencephalography (EEG) Data
Challenges in using EEG data
• Compact comfortable headsets
• Noisy Data
• Online, fully automated algorithms
• Hardware implementations
3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 10
Challenges in using EEG data
• Compact comfortable headsets
• Noisy Data
• Online, fully automated algorithms
• Hardware implementations
3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 11
Challenges in using EEG data
• Compact comfortable headsets
• Noisy Data
• Online, fully automated algorithms
• Hardware implementations
3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 12
EEG Artifacts
• Artifacts: Unwanted electrical activity arising from sources
other than the brain
• Types of Artifacts
– Biological Artifacts
• Ocular artifacts
• Muscular artifacts
• Cardiac artifacts
– External artifacts
• Electrode motion
• External device artifacts
3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 13
EEG Artifacts
• Artifacts: Unwanted electrical activity arising from sources
other than the brain
• Types of Artifacts
– Biological Artifacts
• Ocular artifacts
• Muscular artifacts
• Cardiac artifacts
– External artifacts
• Electrode motion
• External device artifacts
3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 14
Eye Blink Artifact
EEG Artifacts
• Artifacts: Unwanted electrical activity arising from sources other
than the brain
• Types of Artifacts
– Biological Artifacts
– External Artifacts
• Why is it important to remove them?
– Increase the chance of false alarms in seizure detection (Seneviratne et al.
2013)
– They can also alter the shape of neurological events causing unintentional
control of BCI systems (Vaughan et al. 2003).
3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 15
EEG Artifacts
• Artifacts: Unwanted electrical activity arising from sources other
than the brain
• Types of Artifacts
– Biological Artifacts
– External Artifacts
• Why is it important to remove them?
• Can occurrence of artifacts be avoided?
– The occurrence of external artifacts can be reduced by proper placement
of electrodes, but it is impossible to avoid artifacts of biological origin.
3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 16
Contributions
• Propose a system for artifact removal from EEG data using
weak supervisory information.
• Improve the model by proposing an online, fully automated,
end-to-end system for artifact removal trained using unpaired
training corpora.
• Creating a framework for evaluation of artifact removal
algorithms
• Improving artifact removal using additional annotations
• Generalizing AsymmetricGAN for denoising speech data
3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 17
Contributions
• Propose a system for artifact removal from EEG data using
weak supervisory information.
• Improve the model by proposing an online, fully automated,
end-to-end system for artifact removal trained using unpaired
training corpora.
• Creating a framework for evaluation of artifact removal
algorithms
• Improving artifact removal using additional annotations
• Generalizing AsymmetricGAN for denoising speech data
3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 18
• Introduction To Artifact Removal From EEG data
• ICA And Multi-instance Learning Solution
• Asymmetric Generative Adversarial Network Solution
• Future Work
• Questions
3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 19
Related Work
• Artifact Rejection(Nolan, Whelan, and Reilly 2010)
– Train an artifact detector
– Remove the segments of data where artifacts are present
– Excessive loss of information
• Regression using reference channel (Croft and Barry 2000)
– Use a reference channel like Electrocardiography (ECG), Electrooculography
(EOG) or Electromyography (EMG).
– Fail if a reference signal is not available.
• Adaptive filtering (Sweeney, Ward, and McLoone 2012)
– Generate an artifact signal that is uncorrelated to the EEG signal but correlated
to the reference signal.
– Fail if a reference signal is not available.
3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 20
Related Work
3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 21
• Wavelet transforms (Islam et al. 2017) (Daly et al. 2015)
– Decompose each channel using DWT
– Threshold the low frequency component and reconstruct the signal
– Process each channel separately and could miss important clues for
removing the artifact.
22
Related Work
• Principal component analysis (PCA) (Berg and Scherg 1991)
– One of the initial approaches that used principal component analysis
(PCA) for decomposing the signal and removing artifacts from the
decomposed signal.
– Uses orthogonal transformation to convert a multi channel EEG signal
into a set of linearly uncorrelated variables.
– It has been demonstrated that PCA is unable to separate some
artifactual components from brain signals, especially when they have
similar amplitudes (Fitzgibbon et al. 2007b).
3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 23
Independent component analysis (ICA)
• Most popular technique for artifact removal from EEG data
• ICA is used to recover independent source signals called
components and then the components corresponding to
artifacts are identified.
3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 24
ICA Based Artifact Removal
• Shackman, Alexander J., et al. "Identifying robust and sensitive frequency bands for
interrogating neural oscillations." Neuroimage 51.4 (2010): 1319-1333.
• Fitzgibbon, S. P., et al. "Removal of EEG noise and artifact using blind source separation."
Journal of Clinical Neurophysiology 24.3 (2007): 232-243.
• Joyce, Carrie A., Irina F. Gorodnitsky, and Marta Kutas. "Automatic removal of eye
movement and blink artifacts from EEG data using blind component separation."
Psychophysiology 41.2 (2004): 313-325.
• Jung, Tzyy-Ping, et al. "Removal of eye activity artifacts from visual event-related potentials
in normal and clinical subjects." Clinical Neurophysiology 111.10 (2000): 1745-1758.
• Jung, Tzyy-Ping, et al. "Removing electroencephalographic artifacts by blind source
separation." Psychophysiology 37.02 (2000): 163-178.
• Keren, Alon S., Shlomit Yuval-Greenberg, and Leon Y. Deouell. "Saccadic spike potentials in
gamma-band EEG: characterization, detection and suppression." Neuroimage 49.3 (2010):
2248-2263.
253/2/2019 Artifact Removal From Electroencephalography (EEG) Data
ICA Based Artifact Removal
• Shackman, Alexander J., et al. "Identifying robust and sensitive frequency bands for
interrogating neural oscillations." Neuroimage 51.4 (2010): 1319-1333.
• Fitzgibbon, S. P., et al. "Removal of EEG noise and artifact using blind source separation."
Journal of Clinical Neurophysiology 24.3 (2007): 232-243.
• Joyce, Carrie A., Irina F. Gorodnitsky, and Marta Kutas. "Automatic removal of eye
movement and blink artifacts from EEG data using blind component separation."
Psychophysiology 41.2 (2004): 313-325.
• Jung, Tzyy-Ping, et al. "Removal of eye activity artifacts from visual event-related potentials
in normal and clinical subjects." Clinical Neurophysiology 111.10 (2000): 1745-1758.
• Jung, Tzyy-Ping, et al. "Removing electroencephalographic artifacts by blind source
separation." Psychophysiology 37.02 (2000): 163-178.
• Keren, Alon S., Shlomit Yuval-Greenberg, and Leon Y. Deouell. "Saccadic spike potentials in
gamma-band EEG: characterization, detection and suppression." Neuroimage 49.3 (2010):
2248-2263.
Artifact components need to be manually
identified or supervisory signal is needed for
training
263/2/2019 Artifact Removal From Electroencephalography (EEG) Data
ICA Based Artifact Removal
• Shackman, Alexander J., et al. "Identifying robust and sensitive frequency bands for
interrogating neural oscillations." Neuroimage 51.4 (2010): 1319-1333.
• Fitzgibbon, S. P., et al. "Removal of EEG noise and artifact using blind source separation."
Journal of Clinical Neurophysiology 24.3 (2007): 232-243.
• Joyce, Carrie A., Irina F. Gorodnitsky, and Marta Kutas. "Automatic removal of eye
movement and blink artifacts from EEG data using blind component separation."
Psychophysiology 41.2 (2004): 313-325.
• Jung, Tzyy-Ping, et al. "Removal of eye activity artifacts from visual event-related potentials
in normal and clinical subjects." Clinical Neurophysiology 111.10 (2000): 1745-1758.
• Jung, Tzyy-Ping, et al. "Removing electroencephalographic artifacts by blind source
separation." Psychophysiology 37.02 (2000): 163-178.
• Keren, Alon S., Shlomit Yuval-Greenberg, and Leon Y. Deouell. "Saccadic spike potentials in
gamma-band EEG: characterization, detection and suppression." Neuroimage 49.3 (2010):
2248-2263.
Key Idea: Identify components corresponding to
artifacts using weak supervisory signal
273/2/2019 Artifact Removal From Electroencephalography (EEG) Data
System Architecture
28
Mul$ Channel EEG Signal
Channel 1
Channel 2
Channel 64
.
...
..
ICA
Bag of components
.
..
SAX
Feature
Extrac0on .
..
0 1 0 1 0 0
Mul0-
Instance
Learning
(MIL)
0 0 1 1 0 0
0 0 1 1 1 1
Vector represen$ng
component
Sax string: "aaa" "abc" "aab" "aba" "acb" "acc"
Probability of noise
for each component
Ar$fact/ Not Ar$fact?
• Main Processing Blocks
⎻ Independent Component Analysis
⎻ SAX feature extraction
⎻ Multi Instance Learning
3/2/2019 Artifact Removal From Electroencephalography (EEG) Data
Independent component analysis (ICA)
3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 29
ICA
Origin 1
Origin 2
Origin 3
Mixing
System
Observed 1
Observed 2
Observed 3
Estimated 1
Estimated 2
Estimated 3
Unknown
Independent component analysis (ICA)
3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 30
Source: http://scikit-learn.sourceforge.net/0.6/auto_examples/plot_ica_blind_source_separation.html
Independent component analysis (ICA)
3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 31
ICA
Origin 1
Origin 2
Origin 3
Mixing
System
Observed 1
Observed 2
Observed 3
Estimated 1
Estimated 2
Estimated 3
Unknown
Independent component analysis (ICA)
3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 32
ICA
Clean EEG
Signal
Artifact
Mixing
System
Observed 1
Observed 2
Observed 3
Denoised EEG
Artifact
Unknown
System Architecture
33
Mul$ Channel EEG Signal
Channel 1
Channel 2
Channel 64
.
...
..
ICA
Bag of components
.
..
SAX
Feature
Extrac0on .
..
0 1 0 1 0 0
Mul0-
Instance
Learning
(MIL)
0 0 1 1 0 0
0 0 1 1 1 1
Vector represen$ng
component
Sax string: "aaa" "abc" "aab" "aba" "acb" "acc"
Probability of noise
for each component
Ar$fact/ Not Ar$fact?
• Main Processing Blocks
⎻ Independent Component Analysis
⎻ SAX feature extraction
⎻ Multi Instance Learning
3/2/2019 Artifact Removal From Electroencephalography (EEG) Data
Slide by Eamonn Keogh(http://www.cs.ucr.edu/~eamonn/SAX.ppt)
34
How do we obtain SAX features?
0
00 20 40 60 80 100 120
bb
b
a
c
c
c
a
baabccbc
0
20
20 40 60 80 100 120
3/2/2019 Artifact Removal From Electroencephalography (EEG) Data
System Architecture
35
Mul$ Channel EEG Signal
Channel 1
Channel 2
Channel 64
.
...
..
ICA
Bag of components
.
..
SAX
Feature
Extrac0on .
..
0 1 0 1 0 0
Mul0-
Instance
Learning
(MIL)
0 0 1 1 0 0
0 0 1 1 1 1
Vector represen$ng
component
Sax string: "aaa" "abc" "aab" "aba" "acb" "acc"
Probability of noise
for each component
Ar$fact/ Not Ar$fact?
• Main Processing Blocks
⎻ Independent Component Analysis
⎻ SAX feature extraction
⎻ Multi Instance Learning
3/2/2019 Artifact Removal From Electroencephalography (EEG) Data
Multi-Instance Learning
36
Multi-Instance Learning
• A bag is labelled positive if it has at least one positive instance
• Goal: Predict bag label and instance labels
Bag
Positive/Negative Bag?
373/2/2019 Artifact Removal From Electroencephalography (EEG) Data
Multi-Instance Learning
• A bag is labelled positive if it has at least one positive instance
• Goal: Predict bag label and instance label
Bag
Positive/Negative Instance?
Positive/Negative Instance?
Positive/Negative Instance?
Positive/Negative Instance?
383/2/2019 Artifact Removal From Electroencephalography (EEG) Data
System Architecture
39
Mul$ Channel EEG Signal
Channel 1
Channel 2
Channel 64
.
...
..
ICA
Bag of components
.
..
SAX
Feature
Extrac0on .
..
0 1 0 1 0 0
Mul0-
Instance
Learning
(MIL)
0 0 1 1 0 0
0 0 1 1 1 1
Vector represen$ng
component
Sax string: "aaa" "abc" "aab" "aba" "acb" "acc"
Probability of noise
for each component
Ar$fact/ Not Ar$fact?
• Main Processing Blocks
⎻ Independent Component Analysis
⎻ SAX feature extraction
⎻ Multi Instance Learning
3/2/2019 Artifact Removal From Electroencephalography (EEG) Data
EEG Dataset Generated by the US Army Research
Laboratory
40
EEG when participant is
not performing any activity
• 17 subjects
• Artifacts
o Sync pulse
o Clench jaw
o Move jaw vertically
o Blink (no squinting!)
o Move eyes left, then back
to center
o Move eyes up, then back
to center
o Raise and lower eyebrows
o Rotate head side-to-side
o Shrug shoulders
o Rotate torso (hips) EEG when participant is
raising and lowering eyebrows
3/2/2019 Artifact Removal From Electroencephalography (EEG) Data
Results
41
Algorithm Kernel Accuracy
Normalized Set Kernel Linear 95.2%
MISVM Linear 73.2%
miSVM Linear 70.0%
MISVM Quadratic Kernel 81.6%
miSVM Quadratic Kernel 66.0%
3/2/2019 Artifact Removal From Electroencephalography (EEG) Data
Results
3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 42
Patient ID LOSO Accuracy Personalized Accuracy
1 72.54901961 98.04(3.39)
2 76.47058824 96.08(3.39)
3 76.47058824 92.16(6.79)
4 82.35294118 98.04(3.39)
5 84.31372549 98.04(3.39)
6 31.37254902 94.12(0)
7 94.11764706 100(0)
8 100 100(0)
9 98.03921569 100(0)
10 66.66666667 90.2(6.8)
11 98.03921569 100(0)
12 80.39215686 94.12(10.19)
13 88.23529412 100(0)
14 49.01960784 100(0)
15 66.66666667 90.2(12.25)
16 54.90196078 100(0)
17 90.19607843 100(0)
Average 77.0472895 97.12
Accuracy
3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 43
Patient ID LOSO Accuracy Personalized Accuracy
1 72.54901961 98.04(3.39)
2 76.47058824 96.08(3.39)
3 76.47058824 92.16(6.79)
4 82.35294118 98.04(3.39)
5 84.31372549 98.04(3.39)
6 31.37254902 94.12(0)
7 94.11764706 100(0)
8 100 100(0)
9 98.03921569 100(0)
10 66.66666667 90.2(6.8)
11 98.03921569 100(0)
12 80.39215686 94.12(10.19)
13 88.23529412 100(0)
14 49.01960784 100(0)
15 66.66666667 90.2(12.25)
16 54.90196078 100(0)
17 90.19607843 100(0)
Average 77.0472895 97.12
Patient ID Accuracy
1 100(0)
2 89(7.3)
3 100(0)
4 99(2.8)
5 100((0)
6 97(5.5)
7 97(5.5)
Average 98(5.5)
Pooled 97(1.9)
Bootstrap 14(17.7)
LOSO 75
Lawhern, Vernon, W. David Hairston, Kaleb McDowell,
Marissa Westerfield, and Kay Robbins. "Detection and
classification of subject-generated artifacts in EEG
signals using autoregressive models." Journal of
neuroscience methods 208, no. 2 (2012): 181-189.
Summary
• We presented an EEG artifact removal system, using ICA, SAX
feature extraction, and Multi-Instance Learning algorithms. The
proposed system uses a weak supervisory signal to indicate
that some noise is occurring, but not what the source of the
noise is or how it is manifested in the EEG signal. We optimize
the hyper-parameters of the system to reduce the execution
time of the system while maintaining accuracy.
45
Jafari, Ali, Sunil Gandhi, Sri Harsha Konuru, W. David Hairston, Tim Oates, and Tinoosh Mohsenin. "An EEG artifact
identification embedded system using ICA and multi-instance learning." In Circuits and Systems (ISCAS), 2017 IEEE
International Symposium on, pp. 1-4. IEEE, 2017.
3/2/2019 Artifact Removal From Electroencephalography (EEG) Data
Limitations
• Independent component analysis needs large number of
samples to converge. ICA has high computational complexity
and large memory requirements, making it unsuitable for real-
time applications.
• Each subsystem has its own hyperparameters and tuning them
jointly is a challenging task.
463/2/2019 Artifact Removal From Electroencephalography (EEG) Data
Limitations
• Independent component analysis needs large number of
samples to converge. ICA has high computational complexity
and large memory requirements, making it unsuitable for real-
time applications.
• Each subsystem has its own hyperparameters and tuning them
jointly is a challenging task. Also,
47
Can we learn an end to end system to generate
clean EEG from noisy EEG using deep neural
networks
3/2/2019 Artifact Removal From Electroencephalography (EEG) Data
• Introduction To Artifact Removal From EEG data
• ICA And Multi-instance Learning Solution
• Asymmetric Generative Adversarial Network Solution
• Future Work
• Questions
3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 48
Background
• Generative Adversarial network (GAN)
• CycleGAN: Unpaired Image-to-Image Translation using Cycle-
Consistent Adversarial Networks
3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 49
Generative Adversarial Networks
• Generator network: try to fool the discriminator by generating real-looking images
• Discriminator network: try to distinguish between real and fake images
real or fake?
Discriminator
z
G(z)
D
Generator
G
3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 50
[Goodfellow et al. 2014]
real
z G(z)
DG
D
x
fake
51
CycleGAN: Unpaired Image-to-image translation
using cycle-consistent adversarial networks
52
…
[Zhu*, Park*, Isola, Efros on arxiv]
……
X
Dx
Y
Dy
G
F
Cycle GAN
53
Cycle-consistency Loss
Backward cycle loss: 𝐺 𝐹 𝑦 − 𝑦 1
Forward cycle loss: F G x − x 1
G(x) F(G x )x F(y) G(F x )𝑦
54
CycleGAN: Unpaired Image-to-image translation
using cycle-consistent adversarial networks
EEG Data
EEG Data without
Artifacts 55
CycleGAN Model
A B
G_B
G_A
Da Db
3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 56
CycleGAN Model
A B
G_B
G_A
Da Db
• Problems
– Network G_A does not have
access to predicted noise
signal to reconstruct A
– The network is penalized for
not adding noise that is the
same as predicted noise
3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 57
CycleGAN Model
A B
G_B
G_A
Da Db
A
A’
G_B
G_A
3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 58
AsymmetricGAN Model
A B
G_B
G_A
Da Db
We preserve the noise in the signal and use it for reconstruction of the original signal
3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 59
A BN
G_B
G_NG_N
+
Da Db
AsymmetricGAN Model
A B
G_B
G_A
Da Db
+
3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 60
AsymmetricGAN Model
3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 61
G_N is a function that extracts noise if the input is a noisy signal A. It generates noise if the input is a clean signal B.
Generator and discriminator network architecture
3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 62
Generator and discriminator network architecture
3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 63
• Network is independent of the
relative ordering of the
channels
• Channel indices do not
correspond to the spatial
locations of the electrodes.
• Filter Size:
• 1D convolution with filter
size K
• 2D convolution with filter
size N ✕ k
Qi, Charles R., Hao Su, Kaichun Mo, and Leonidas J. Guibas.
"Pointnet: Deep learning on point sets for 3d classification and
segmentation." Proc. Computer Vision and Pattern
Recognition (CVPR), IEEE 1, no. 2 (2017): 4.
Evaluation
• Evaluation of artifact removal from EEG data is difficult as
ground truth signal does not exist.
• We just have set of clean EEG signals and noisy EEG signals. We
don’t have clean version of noisy EEG.
• Visualizing and understanding EEG data is a time consuming
task.
• To validate the model we create a simpler synthetic dataset.
3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 64
Synthetic Dataset
• Clean signal is a linear combination of a sine and a square wave
• Noisy signal is linear combination of sine, square and sawtooth
waves
• The period of the sine and square waves is randomly selected
between 2 and 5. Sawtooth has fixed period of 6
• Number of samples in each time series is 1000
• Training set: 4000 Signals
• Validation set: 1000 Signals
• Test set: 100 Signals
3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 65
Synthetic Dataset
3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 66
(e) True Sources
(c) Signal cleaned by Asymmetric GAN
(a) Signal with Artifact (b) Clean Signal
(d) Noise Signal predicted by Asymmetric GAN
Synthetic Dataset
3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 67
(e) True Sources
(c) Signal cleaned by Asymmetric GAN
(a) Signal with Artifact (b) Clean Signal
(d) Noise Signal predicted by Asymmetric GAN
The MSE between the ground truth clean signal and
denoised signal is 0.0406. The mean MSE error for entire
test set is 0.0387 and standard deviation is 0.0043.
EEG Dataset Generated by the US Army Research
Laboratory
• During collection of this dataset, participants were told to not move and look
straight at the computer screen for the collection of clean data.
• Despite the instruction to the patients, we noticed that there were artifacts even
in “clean” data. We also noticed that some “noisy” EEG did not contain the
corresponding artifact.
• Manually annotating all artifacts from all channels for all artifacts is a time
consuming task. So in this work, we focus on ocular artifacts in the Fp1 electrode
of the frontal region.
• Remove all patients that have more than two ocular artifacts in the clean data
and do not have artifacts in the region of eyebrow raising.
• In the resulting dataset, we have 4 patients with clean data and 10 patients with
noisy data.
• Each patient’s clean data has 4836 samples and noisy data has 420354 samples.
We use this manually annotated data in all experiments below.
3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 68
Evaluation
• Qualitative Results
• Using a detector
3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 69
Qualitative Results
(a) (b)
3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 70
Evaluation by detection
3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 71
• We use artifact detection as a way of measuring the
performance of artifact removal.
• We first train artifact detection and artifact removal and use
the artifact detector to classify every window in EEG data
denoised by the artifact removal algorithm.
• The error is given by the percent of windows where an artifact
was detected in denoised EEG data.
Dataset Division
3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 72
0
0.5
1
1.5
2
2.5
3
3.5
4
4.5
Clean Noisy Clean Noisy Clean Noisy Clean Noisy
Train Test Train Test
Artifact Detector Artifact Removal
Evaluation by detection
• Artifact Detector accuracy: 97.39%
• Denoised signal was classified as clean by artifact detector
72.37% times.
3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 73
Summary
• We presented an online, fully automated, end-to-end system
for denoising EEG data. Our system for denoising is trained
using unpaired training corpora. It does not need any
information about the source of the noise or how it is
manifested in the EEG signal. We created a synthetic dataset
and used it to validate our network. We also used our system
to remove artifacts from existing EEG dataset
3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 74
Sunil Gandhi, Tim Oates, Tinoosh Mohsenin, Dave Hairston. ”Denoising Time Series Data Using Asymmetric Generative
Adversarial Networks.” In Pacific-Asia Conference on Knowledge Discovery and Data Mining, 2018.
• Introduction To Artifact Removal From EEG data
• ICA And Multi-instance Learning Solution
• Asymmetric Generative Adversarial Network Solution
• Future Work
• Questions
3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 75
Future Work
• Creating a framework for evaluation of artifact removal
algorithms
• Removing multiple artifacts using AsymmetricGAN
• Improving artifact removal using additional annotations
• Generalizing AsymmetricGAN for speech data
3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 76
Evaluation of artifact removal algorithms
• Evaluating artifact removal algorithms is important before they
can be used in clinical contexts.
• But, it is challenging because of a lack of availability of ground
truth clean EEG signals.
3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 77
Evaluation of artifact removal algorithms
• Evaluating artifact removal algorithms is important before they
can be used in clinical contexts.
• But, it is challenging because of a lack of availability of ground
truth clean EEG signals.
3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 78
The performance evaluation of artifact removal methods found in the literature is
always problematic. It can be done either by visually by expert which is subjective
or by synthetic/semi-synthetic data (but uncertainty of reconstructed data whether
perfectly realistic or not). Since there is neither any ground truth data available nor
any universal or standard quantitative metric(s) used in the literature that can
capture both amount of artifact removal and distortion. … Therefore, it is not fair
to tell which performs best based on the study.
Islam, Md Kafiul, Amir Rastegarnia, and Zhi Yang. "Methods for artifact detection and removal from scalp EEG: A
review." Neurophysiologie Clinique/Clinical Neurophysiology 46, no. 4-5 (2016): 287-305.
Evaluation of artifact removal algorithms
• Qualitative Evaluation
• Evaluation using simulated data
• Evaluation using detection
• Correlating with the reference signal
3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 79
Evaluation of artifact removal algorithms
• Qualitative Evaluation
– Subjective
– Does not give quantitative metric
3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 80
Evaluation of artifact removal algorithms
• Qualitative Evaluation
• Evaluation using simulated data
– Allows usage of metrics like signal-to-noise ratio (SNR) and
normalized mean squared error to compare deviation between the
denoised signal and clean signal.
– Training and evaluation can be performed on large datasets
– But it is very difficult to replicate all the characteristics of EEG data
like synchronization among channels, time-locking to ERPs and
contamination by different types of artifacts in a realistic manner.
3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 81
Evaluation of artifact removal algorithms
• Qualitative Evaluation
• Evaluation using simulated data
• Evaluation using detection
– Use artifact detector to check if artifacts are present in the denoised signal.
– Can be used to evaluate denoising of real EEG data
– Automated, not subjective and gives a quantitative measure to compare several
artifact removal algorithms.
– Usefulness of this metric is dependent on the performance of the artifact
detector.
– Does not give a measure of the sensitivity (whether the method removes the
artifacts) and specificity (whether it preserves EEG signals) of the artifact
removal algorithm.
3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 82
Evaluation of artifact removal algorithms
• Qualitative Evaluation
• Evaluation using simulated data
• Evaluation using detection
• Correlating with the reference signal
– Cannot be performed if a reference signal for the corresponding
artifact is not available.
– Does not give a measure of the sensitivity (whether the method
removes the artifacts) and specificity (whether it preserves EEG
signals) of the artifact removal algorithm.
3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 83
Evaluation of artifact removal algorithms
• Evaluation Methods
– Qualitative Evaluation
– Evaluation using simulated data
– Evaluation using detection
– Correlating with the reference signal
• Multiple evaluation methods with several variants of each of these methods exist in
literature.
• Each of these methods has their strengths and limitations. This makes performance
comparison of these methods hard.
• In future we plan to
– Standardize the evaluation mechanism
– Create a tool for evaluating artifact removal algorithms using all of the above approaches.
– Comparing our approach with existing methods using the tool
3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 84
Future Work
• Creating a framework for evaluation of artifact removal
algorithms
• Removing multiple artifacts using AsymmetricGAN
• Improving artifact removal using additional annotations
• Generalizing AsymmetricGAN for speech data
3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 85
Removing multiple artifacts using AsymmetricGAN
• Having separate algorithms for removal of different types of
artifacts makes the preprocessing step inefficient.
• In future, we plan to adapt the AsymmetricGAN architecture for
removing multiple types of artifacts.
3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 86
G_B
G_NG_N
+
Da Db
MAA B
OA
Future Work
• Creating a framework for evaluation of artifact removal
algorithms
• Removing multiple artifacts using AsymmetricGAN
• Improving artifact removal using additional annotations
• Generalizing AsymmetricGAN for speech data
3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 87
Improving artifact removal using additional
annotations
• Modify AsymmetricGAN to utilize
– artifact type
– location of artifact
– reference channel.
• We will study differences in performance because of each type
of additional data.
• This will give insight on which annotations are most helpful for
artifact removal algorithms, thus helping creation of future
datasets.
3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 88
Future Work
• Creating a framework for evaluation of artifact removal
algorithms
• Removing multiple artifacts using AsymmetricGAN
• Improving artifact removal using additional annotations
• Generalizing AsymmetricGAN for speech data
3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 89
Generalizing AsymmetricGAN for speech data
• Speech denoising produces noise-free speech signals from noisy recordings
• Applications
– Recognizing speech in environments with background noise like car driving.
– Reduce discomfort and increase understanding of the speech for people wearing
hearing aids
• Easier to evaluate
• Evaluate generalizability of asymmetricGAN across domains
• Datasets:
– Voice Bank corpus: Speech data for 400 sentences from 28 speakers each.
– Demand dataset: Real world background noise of 18 diverse environments
3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 90
Questions?
3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 91

More Related Content

What's hot

Bioimpedance overview
Bioimpedance overview Bioimpedance overview
Bioimpedance overview ES-Teck India
 
UG thesis presentation
UG thesis presentationUG thesis presentation
UG thesis presentationLuke Zhao
 
A Biomedical Smart Sensor for Visually impaired
A Biomedical Smart Sensor for Visually impairedA Biomedical Smart Sensor for Visually impaired
A Biomedical Smart Sensor for Visually impairedDinesh Mv
 
Sleep quality prediction from wearable data using deep learning
Sleep quality prediction from wearable data using deep learningSleep quality prediction from wearable data using deep learning
Sleep quality prediction from wearable data using deep learningLuis Fernandez Luque
 
AN ELECTRICAL IMPEDANCE TOMOGRAPHY SYSTEM FOR THYROID GLAND WITH A TINY ELECT...
AN ELECTRICAL IMPEDANCE TOMOGRAPHY SYSTEM FOR THYROID GLAND WITH A TINY ELECT...AN ELECTRICAL IMPEDANCE TOMOGRAPHY SYSTEM FOR THYROID GLAND WITH A TINY ELECT...
AN ELECTRICAL IMPEDANCE TOMOGRAPHY SYSTEM FOR THYROID GLAND WITH A TINY ELECT...ijbesjournal
 

What's hot (6)

Bioimpedance overview
Bioimpedance overview Bioimpedance overview
Bioimpedance overview
 
UG thesis presentation
UG thesis presentationUG thesis presentation
UG thesis presentation
 
A Biomedical Smart Sensor for Visually impaired
A Biomedical Smart Sensor for Visually impairedA Biomedical Smart Sensor for Visually impaired
A Biomedical Smart Sensor for Visually impaired
 
Ijetcas14 323
Ijetcas14 323Ijetcas14 323
Ijetcas14 323
 
Sleep quality prediction from wearable data using deep learning
Sleep quality prediction from wearable data using deep learningSleep quality prediction from wearable data using deep learning
Sleep quality prediction from wearable data using deep learning
 
AN ELECTRICAL IMPEDANCE TOMOGRAPHY SYSTEM FOR THYROID GLAND WITH A TINY ELECT...
AN ELECTRICAL IMPEDANCE TOMOGRAPHY SYSTEM FOR THYROID GLAND WITH A TINY ELECT...AN ELECTRICAL IMPEDANCE TOMOGRAPHY SYSTEM FOR THYROID GLAND WITH A TINY ELECT...
AN ELECTRICAL IMPEDANCE TOMOGRAPHY SYSTEM FOR THYROID GLAND WITH A TINY ELECT...
 

Similar to Two Algorithms Weakly Supervised Denoising EEG Data

Automated_EEG_artifact_elimination_by_applying_mac.pdf
Automated_EEG_artifact_elimination_by_applying_mac.pdfAutomated_EEG_artifact_elimination_by_applying_mac.pdf
Automated_EEG_artifact_elimination_by_applying_mac.pdfmurali926139
 
Study and analysis of motion artifacts for ambulatory electroencephalography
Study and analysis of motion artifacts for ambulatory  electroencephalographyStudy and analysis of motion artifacts for ambulatory  electroencephalography
Study and analysis of motion artifacts for ambulatory electroencephalographyIJECEIAES
 
CLASSIFICATION OF ELECTROENCEPHALOGRAM SIGNALS USING XGBOOST ALGORITHM AND SU...
CLASSIFICATION OF ELECTROENCEPHALOGRAM SIGNALS USING XGBOOST ALGORITHM AND SU...CLASSIFICATION OF ELECTROENCEPHALOGRAM SIGNALS USING XGBOOST ALGORITHM AND SU...
CLASSIFICATION OF ELECTROENCEPHALOGRAM SIGNALS USING XGBOOST ALGORITHM AND SU...IRJET Journal
 
PhD Oral Defense of Md Kafiul Islam on "ARTIFACT CHARACTERIZATION, DETECTION ...
PhD Oral Defense of Md Kafiul Islam on "ARTIFACT CHARACTERIZATION, DETECTION ...PhD Oral Defense of Md Kafiul Islam on "ARTIFACT CHARACTERIZATION, DETECTION ...
PhD Oral Defense of Md Kafiul Islam on "ARTIFACT CHARACTERIZATION, DETECTION ...Md Kafiul Islam
 
FrB18_2_Krishnamoorthy_Venkatasubramanian.pptx
FrB18_2_Krishnamoorthy_Venkatasubramanian.pptxFrB18_2_Krishnamoorthy_Venkatasubramanian.pptx
FrB18_2_Krishnamoorthy_Venkatasubramanian.pptxVenkatasubramanianKr8
 
Brain Computer Interface ppt
Brain Computer Interface pptBrain Computer Interface ppt
Brain Computer Interface pptAjay George
 
A New Approach to Denoising EEG Signals - Merger of Translation Invariant Wav...
A New Approach to Denoising EEG Signals - Merger of Translation Invariant Wav...A New Approach to Denoising EEG Signals - Merger of Translation Invariant Wav...
A New Approach to Denoising EEG Signals - Merger of Translation Invariant Wav...CSCJournals
 
IRJET-Denoising and Eye Movement Correction in EEG Recordings using Discrete ...
IRJET-Denoising and Eye Movement Correction in EEG Recordings using Discrete ...IRJET-Denoising and Eye Movement Correction in EEG Recordings using Discrete ...
IRJET-Denoising and Eye Movement Correction in EEG Recordings using Discrete ...IRJET Journal
 
Prediction Model for Emotion Recognition Using EEG
Prediction Model for Emotion Recognition Using EEGPrediction Model for Emotion Recognition Using EEG
Prediction Model for Emotion Recognition Using EEGIRJET Journal
 
Smart Home for Paralyzed Aid
Smart Home for Paralyzed AidSmart Home for Paralyzed Aid
Smart Home for Paralyzed AidIRJET Journal
 
Brain computer interface based
Brain computer interface basedBrain computer interface based
Brain computer interface basedMohammed Saeed
 
Denoising of EEG Signals for Analysis of Brain Disorders: A Review
Denoising of EEG Signals for Analysis of Brain Disorders: A ReviewDenoising of EEG Signals for Analysis of Brain Disorders: A Review
Denoising of EEG Signals for Analysis of Brain Disorders: A ReviewIRJET Journal
 
Invited talk at IBRO UIU EEG Signal Processing
Invited talk at IBRO UIU EEG Signal ProcessingInvited talk at IBRO UIU EEG Signal Processing
Invited talk at IBRO UIU EEG Signal ProcessingMd Kafiul Islam
 
Epilepsy Prediction using Machine Learning
Epilepsy Prediction using Machine LearningEpilepsy Prediction using Machine Learning
Epilepsy Prediction using Machine LearningIRJET Journal
 
IRJET- Deep Neural Network for the Automated Detection and Diagnosis of Seizu...
IRJET- Deep Neural Network for the Automated Detection and Diagnosis of Seizu...IRJET- Deep Neural Network for the Automated Detection and Diagnosis of Seizu...
IRJET- Deep Neural Network for the Automated Detection and Diagnosis of Seizu...IRJET Journal
 
A Comparative Study of Machine Learning Algorithms for EEG Signal Classification
A Comparative Study of Machine Learning Algorithms for EEG Signal ClassificationA Comparative Study of Machine Learning Algorithms for EEG Signal Classification
A Comparative Study of Machine Learning Algorithms for EEG Signal Classificationsipij
 
A COMPARATIVE STUDY OF MACHINE LEARNING ALGORITHMS FOR EEG SIGNAL CLASSIFICATION
A COMPARATIVE STUDY OF MACHINE LEARNING ALGORITHMS FOR EEG SIGNAL CLASSIFICATIONA COMPARATIVE STUDY OF MACHINE LEARNING ALGORITHMS FOR EEG SIGNAL CLASSIFICATION
A COMPARATIVE STUDY OF MACHINE LEARNING ALGORITHMS FOR EEG SIGNAL CLASSIFICATIONsipij
 

Similar to Two Algorithms Weakly Supervised Denoising EEG Data (20)

Automated_EEG_artifact_elimination_by_applying_mac.pdf
Automated_EEG_artifact_elimination_by_applying_mac.pdfAutomated_EEG_artifact_elimination_by_applying_mac.pdf
Automated_EEG_artifact_elimination_by_applying_mac.pdf
 
Study and analysis of motion artifacts for ambulatory electroencephalography
Study and analysis of motion artifacts for ambulatory  electroencephalographyStudy and analysis of motion artifacts for ambulatory  electroencephalography
Study and analysis of motion artifacts for ambulatory electroencephalography
 
CLASSIFICATION OF ELECTROENCEPHALOGRAM SIGNALS USING XGBOOST ALGORITHM AND SU...
CLASSIFICATION OF ELECTROENCEPHALOGRAM SIGNALS USING XGBOOST ALGORITHM AND SU...CLASSIFICATION OF ELECTROENCEPHALOGRAM SIGNALS USING XGBOOST ALGORITHM AND SU...
CLASSIFICATION OF ELECTROENCEPHALOGRAM SIGNALS USING XGBOOST ALGORITHM AND SU...
 
PhD Oral Defense of Md Kafiul Islam on "ARTIFACT CHARACTERIZATION, DETECTION ...
PhD Oral Defense of Md Kafiul Islam on "ARTIFACT CHARACTERIZATION, DETECTION ...PhD Oral Defense of Md Kafiul Islam on "ARTIFACT CHARACTERIZATION, DETECTION ...
PhD Oral Defense of Md Kafiul Islam on "ARTIFACT CHARACTERIZATION, DETECTION ...
 
FrB18_2_Krishnamoorthy_Venkatasubramanian.pptx
FrB18_2_Krishnamoorthy_Venkatasubramanian.pptxFrB18_2_Krishnamoorthy_Venkatasubramanian.pptx
FrB18_2_Krishnamoorthy_Venkatasubramanian.pptx
 
Brain Computer Interface ppt
Brain Computer Interface pptBrain Computer Interface ppt
Brain Computer Interface ppt
 
A New Approach to Denoising EEG Signals - Merger of Translation Invariant Wav...
A New Approach to Denoising EEG Signals - Merger of Translation Invariant Wav...A New Approach to Denoising EEG Signals - Merger of Translation Invariant Wav...
A New Approach to Denoising EEG Signals - Merger of Translation Invariant Wav...
 
IRJET-Denoising and Eye Movement Correction in EEG Recordings using Discrete ...
IRJET-Denoising and Eye Movement Correction in EEG Recordings using Discrete ...IRJET-Denoising and Eye Movement Correction in EEG Recordings using Discrete ...
IRJET-Denoising and Eye Movement Correction in EEG Recordings using Discrete ...
 
Prediction Model for Emotion Recognition Using EEG
Prediction Model for Emotion Recognition Using EEGPrediction Model for Emotion Recognition Using EEG
Prediction Model for Emotion Recognition Using EEG
 
Smart Home for Paralyzed Aid
Smart Home for Paralyzed AidSmart Home for Paralyzed Aid
Smart Home for Paralyzed Aid
 
Priyaseminar presentation
Priyaseminar presentationPriyaseminar presentation
Priyaseminar presentation
 
Presentation1
Presentation1Presentation1
Presentation1
 
Brain computer interface based
Brain computer interface basedBrain computer interface based
Brain computer interface based
 
Kh3517801787
Kh3517801787Kh3517801787
Kh3517801787
 
Denoising of EEG Signals for Analysis of Brain Disorders: A Review
Denoising of EEG Signals for Analysis of Brain Disorders: A ReviewDenoising of EEG Signals for Analysis of Brain Disorders: A Review
Denoising of EEG Signals for Analysis of Brain Disorders: A Review
 
Invited talk at IBRO UIU EEG Signal Processing
Invited talk at IBRO UIU EEG Signal ProcessingInvited talk at IBRO UIU EEG Signal Processing
Invited talk at IBRO UIU EEG Signal Processing
 
Epilepsy Prediction using Machine Learning
Epilepsy Prediction using Machine LearningEpilepsy Prediction using Machine Learning
Epilepsy Prediction using Machine Learning
 
IRJET- Deep Neural Network for the Automated Detection and Diagnosis of Seizu...
IRJET- Deep Neural Network for the Automated Detection and Diagnosis of Seizu...IRJET- Deep Neural Network for the Automated Detection and Diagnosis of Seizu...
IRJET- Deep Neural Network for the Automated Detection and Diagnosis of Seizu...
 
A Comparative Study of Machine Learning Algorithms for EEG Signal Classification
A Comparative Study of Machine Learning Algorithms for EEG Signal ClassificationA Comparative Study of Machine Learning Algorithms for EEG Signal Classification
A Comparative Study of Machine Learning Algorithms for EEG Signal Classification
 
A COMPARATIVE STUDY OF MACHINE LEARNING ALGORITHMS FOR EEG SIGNAL CLASSIFICATION
A COMPARATIVE STUDY OF MACHINE LEARNING ALGORITHMS FOR EEG SIGNAL CLASSIFICATIONA COMPARATIVE STUDY OF MACHINE LEARNING ALGORITHMS FOR EEG SIGNAL CLASSIFICATION
A COMPARATIVE STUDY OF MACHINE LEARNING ALGORITHMS FOR EEG SIGNAL CLASSIFICATION
 

More from Data Works MD

Data Journalism at The Baltimore Banner
Data Journalism at The Baltimore BannerData Journalism at The Baltimore Banner
Data Journalism at The Baltimore BannerData Works MD
 
Jolt’s Picks - Machine Learning and Major League Baseball Hit Streaks
Jolt’s Picks - Machine Learning and Major League Baseball Hit StreaksJolt’s Picks - Machine Learning and Major League Baseball Hit Streaks
Jolt’s Picks - Machine Learning and Major League Baseball Hit StreaksData Works MD
 
Introducing DataWave
Introducing DataWaveIntroducing DataWave
Introducing DataWaveData Works MD
 
Malware Detection, Enabled by Machine Learning
Malware Detection, Enabled by Machine LearningMalware Detection, Enabled by Machine Learning
Malware Detection, Enabled by Machine LearningData Works MD
 
Using AWS, Terraform, and Ansible to Automate Splunk at Scale
Using AWS, Terraform, and Ansible to Automate Splunk at ScaleUsing AWS, Terraform, and Ansible to Automate Splunk at Scale
Using AWS, Terraform, and Ansible to Automate Splunk at ScaleData Works MD
 
A Day in the Life of a Data Journalist
A Day in the Life of a Data JournalistA Day in the Life of a Data Journalist
A Day in the Life of a Data JournalistData Works MD
 
Robotics and Machine Learning: Working with NVIDIA Jetson Kits
Robotics and Machine Learning: Working with NVIDIA Jetson KitsRobotics and Machine Learning: Working with NVIDIA Jetson Kits
Robotics and Machine Learning: Working with NVIDIA Jetson KitsData Works MD
 
Connect Data and Devices with Apache NiFi
Connect Data and Devices with Apache NiFiConnect Data and Devices with Apache NiFi
Connect Data and Devices with Apache NiFiData Works MD
 
Introduction to Machine Learning
Introduction to Machine LearningIntroduction to Machine Learning
Introduction to Machine LearningData Works MD
 
Data in the City: Analytics and Civic Data in Baltimore
Data in the City: Analytics and Civic Data in BaltimoreData in the City: Analytics and Civic Data in Baltimore
Data in the City: Analytics and Civic Data in BaltimoreData Works MD
 
Exploring Correlation Between Sentiment of Environmental Tweets and the Stock...
Exploring Correlation Between Sentiment of Environmental Tweets and the Stock...Exploring Correlation Between Sentiment of Environmental Tweets and the Stock...
Exploring Correlation Between Sentiment of Environmental Tweets and the Stock...Data Works MD
 
Automated Software Requirements Labeling
Automated Software Requirements LabelingAutomated Software Requirements Labeling
Automated Software Requirements LabelingData Works MD
 
Introduction to Elasticsearch for Business Intelligence and Application Insights
Introduction to Elasticsearch for Business Intelligence and Application InsightsIntroduction to Elasticsearch for Business Intelligence and Application Insights
Introduction to Elasticsearch for Business Intelligence and Application InsightsData Works MD
 
An Asynchronous Distributed Deep Learning Based Intrusion Detection System fo...
An Asynchronous Distributed Deep Learning Based Intrusion Detection System fo...An Asynchronous Distributed Deep Learning Based Intrusion Detection System fo...
An Asynchronous Distributed Deep Learning Based Intrusion Detection System fo...Data Works MD
 
RAPIDS – Open GPU-accelerated Data Science
RAPIDS – Open GPU-accelerated Data ScienceRAPIDS – Open GPU-accelerated Data Science
RAPIDS – Open GPU-accelerated Data ScienceData Works MD
 
Detecting Lateral Movement with a Compute-Intense Graph Kernel
Detecting Lateral Movement with a Compute-Intense Graph KernelDetecting Lateral Movement with a Compute-Intense Graph Kernel
Detecting Lateral Movement with a Compute-Intense Graph KernelData Works MD
 
Predictive Analytics and Neighborhood Health
Predictive Analytics and Neighborhood HealthPredictive Analytics and Neighborhood Health
Predictive Analytics and Neighborhood HealthData Works MD
 
Social Network Analysis Workshop
Social Network Analysis WorkshopSocial Network Analysis Workshop
Social Network Analysis WorkshopData Works MD
 

More from Data Works MD (18)

Data Journalism at The Baltimore Banner
Data Journalism at The Baltimore BannerData Journalism at The Baltimore Banner
Data Journalism at The Baltimore Banner
 
Jolt’s Picks - Machine Learning and Major League Baseball Hit Streaks
Jolt’s Picks - Machine Learning and Major League Baseball Hit StreaksJolt’s Picks - Machine Learning and Major League Baseball Hit Streaks
Jolt’s Picks - Machine Learning and Major League Baseball Hit Streaks
 
Introducing DataWave
Introducing DataWaveIntroducing DataWave
Introducing DataWave
 
Malware Detection, Enabled by Machine Learning
Malware Detection, Enabled by Machine LearningMalware Detection, Enabled by Machine Learning
Malware Detection, Enabled by Machine Learning
 
Using AWS, Terraform, and Ansible to Automate Splunk at Scale
Using AWS, Terraform, and Ansible to Automate Splunk at ScaleUsing AWS, Terraform, and Ansible to Automate Splunk at Scale
Using AWS, Terraform, and Ansible to Automate Splunk at Scale
 
A Day in the Life of a Data Journalist
A Day in the Life of a Data JournalistA Day in the Life of a Data Journalist
A Day in the Life of a Data Journalist
 
Robotics and Machine Learning: Working with NVIDIA Jetson Kits
Robotics and Machine Learning: Working with NVIDIA Jetson KitsRobotics and Machine Learning: Working with NVIDIA Jetson Kits
Robotics and Machine Learning: Working with NVIDIA Jetson Kits
 
Connect Data and Devices with Apache NiFi
Connect Data and Devices with Apache NiFiConnect Data and Devices with Apache NiFi
Connect Data and Devices with Apache NiFi
 
Introduction to Machine Learning
Introduction to Machine LearningIntroduction to Machine Learning
Introduction to Machine Learning
 
Data in the City: Analytics and Civic Data in Baltimore
Data in the City: Analytics and Civic Data in BaltimoreData in the City: Analytics and Civic Data in Baltimore
Data in the City: Analytics and Civic Data in Baltimore
 
Exploring Correlation Between Sentiment of Environmental Tweets and the Stock...
Exploring Correlation Between Sentiment of Environmental Tweets and the Stock...Exploring Correlation Between Sentiment of Environmental Tweets and the Stock...
Exploring Correlation Between Sentiment of Environmental Tweets and the Stock...
 
Automated Software Requirements Labeling
Automated Software Requirements LabelingAutomated Software Requirements Labeling
Automated Software Requirements Labeling
 
Introduction to Elasticsearch for Business Intelligence and Application Insights
Introduction to Elasticsearch for Business Intelligence and Application InsightsIntroduction to Elasticsearch for Business Intelligence and Application Insights
Introduction to Elasticsearch for Business Intelligence and Application Insights
 
An Asynchronous Distributed Deep Learning Based Intrusion Detection System fo...
An Asynchronous Distributed Deep Learning Based Intrusion Detection System fo...An Asynchronous Distributed Deep Learning Based Intrusion Detection System fo...
An Asynchronous Distributed Deep Learning Based Intrusion Detection System fo...
 
RAPIDS – Open GPU-accelerated Data Science
RAPIDS – Open GPU-accelerated Data ScienceRAPIDS – Open GPU-accelerated Data Science
RAPIDS – Open GPU-accelerated Data Science
 
Detecting Lateral Movement with a Compute-Intense Graph Kernel
Detecting Lateral Movement with a Compute-Intense Graph KernelDetecting Lateral Movement with a Compute-Intense Graph Kernel
Detecting Lateral Movement with a Compute-Intense Graph Kernel
 
Predictive Analytics and Neighborhood Health
Predictive Analytics and Neighborhood HealthPredictive Analytics and Neighborhood Health
Predictive Analytics and Neighborhood Health
 
Social Network Analysis Workshop
Social Network Analysis WorkshopSocial Network Analysis Workshop
Social Network Analysis Workshop
 

Recently uploaded

Handwritten Text Recognition for manuscripts and early printed texts
Handwritten Text Recognition for manuscripts and early printed textsHandwritten Text Recognition for manuscripts and early printed texts
Handwritten Text Recognition for manuscripts and early printed textsMaria Levchenko
 
Understanding the Laravel MVC Architecture
Understanding the Laravel MVC ArchitectureUnderstanding the Laravel MVC Architecture
Understanding the Laravel MVC ArchitecturePixlogix Infotech
 
Pigging Solutions in Pet Food Manufacturing
Pigging Solutions in Pet Food ManufacturingPigging Solutions in Pet Food Manufacturing
Pigging Solutions in Pet Food ManufacturingPigging Solutions
 
08448380779 Call Girls In Civil Lines Women Seeking Men
08448380779 Call Girls In Civil Lines Women Seeking Men08448380779 Call Girls In Civil Lines Women Seeking Men
08448380779 Call Girls In Civil Lines Women Seeking MenDelhi Call girls
 
The 7 Things I Know About Cyber Security After 25 Years | April 2024
The 7 Things I Know About Cyber Security After 25 Years | April 2024The 7 Things I Know About Cyber Security After 25 Years | April 2024
The 7 Things I Know About Cyber Security After 25 Years | April 2024Rafal Los
 
From Event to Action: Accelerate Your Decision Making with Real-Time Automation
From Event to Action: Accelerate Your Decision Making with Real-Time AutomationFrom Event to Action: Accelerate Your Decision Making with Real-Time Automation
From Event to Action: Accelerate Your Decision Making with Real-Time AutomationSafe Software
 
Install Stable Diffusion in windows machine
Install Stable Diffusion in windows machineInstall Stable Diffusion in windows machine
Install Stable Diffusion in windows machinePadma Pradeep
 
My Hashitalk Indonesia April 2024 Presentation
My Hashitalk Indonesia April 2024 PresentationMy Hashitalk Indonesia April 2024 Presentation
My Hashitalk Indonesia April 2024 PresentationRidwan Fadjar
 
Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...
Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...
Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...shyamraj55
 
Integration and Automation in Practice: CI/CD in Mule Integration and Automat...
Integration and Automation in Practice: CI/CD in Mule Integration and Automat...Integration and Automation in Practice: CI/CD in Mule Integration and Automat...
Integration and Automation in Practice: CI/CD in Mule Integration and Automat...Patryk Bandurski
 
Presentation on how to chat with PDF using ChatGPT code interpreter
Presentation on how to chat with PDF using ChatGPT code interpreterPresentation on how to chat with PDF using ChatGPT code interpreter
Presentation on how to chat with PDF using ChatGPT code interpreternaman860154
 
WhatsApp 9892124323 ✓Call Girls In Kalyan ( Mumbai ) secure service
WhatsApp 9892124323 ✓Call Girls In Kalyan ( Mumbai ) secure serviceWhatsApp 9892124323 ✓Call Girls In Kalyan ( Mumbai ) secure service
WhatsApp 9892124323 ✓Call Girls In Kalyan ( Mumbai ) secure servicePooja Nehwal
 
Factors to Consider When Choosing Accounts Payable Services Providers.pptx
Factors to Consider When Choosing Accounts Payable Services Providers.pptxFactors to Consider When Choosing Accounts Payable Services Providers.pptx
Factors to Consider When Choosing Accounts Payable Services Providers.pptxKatpro Technologies
 
04-2024-HHUG-Sales-and-Marketing-Alignment.pptx
04-2024-HHUG-Sales-and-Marketing-Alignment.pptx04-2024-HHUG-Sales-and-Marketing-Alignment.pptx
04-2024-HHUG-Sales-and-Marketing-Alignment.pptxHampshireHUG
 
08448380779 Call Girls In Greater Kailash - I Women Seeking Men
08448380779 Call Girls In Greater Kailash - I Women Seeking Men08448380779 Call Girls In Greater Kailash - I Women Seeking Men
08448380779 Call Girls In Greater Kailash - I Women Seeking MenDelhi Call girls
 
FULL ENJOY 🔝 8264348440 🔝 Call Girls in Diplomatic Enclave | Delhi
FULL ENJOY 🔝 8264348440 🔝 Call Girls in Diplomatic Enclave | DelhiFULL ENJOY 🔝 8264348440 🔝 Call Girls in Diplomatic Enclave | Delhi
FULL ENJOY 🔝 8264348440 🔝 Call Girls in Diplomatic Enclave | Delhisoniya singh
 
Pigging Solutions Piggable Sweeping Elbows
Pigging Solutions Piggable Sweeping ElbowsPigging Solutions Piggable Sweeping Elbows
Pigging Solutions Piggable Sweeping ElbowsPigging Solutions
 
Key Features Of Token Development (1).pptx
Key  Features Of Token  Development (1).pptxKey  Features Of Token  Development (1).pptx
Key Features Of Token Development (1).pptxLBM Solutions
 
How to Remove Document Management Hurdles with X-Docs?
How to Remove Document Management Hurdles with X-Docs?How to Remove Document Management Hurdles with X-Docs?
How to Remove Document Management Hurdles with X-Docs?XfilesPro
 
How to convert PDF to text with Nanonets
How to convert PDF to text with NanonetsHow to convert PDF to text with Nanonets
How to convert PDF to text with Nanonetsnaman860154
 

Recently uploaded (20)

Handwritten Text Recognition for manuscripts and early printed texts
Handwritten Text Recognition for manuscripts and early printed textsHandwritten Text Recognition for manuscripts and early printed texts
Handwritten Text Recognition for manuscripts and early printed texts
 
Understanding the Laravel MVC Architecture
Understanding the Laravel MVC ArchitectureUnderstanding the Laravel MVC Architecture
Understanding the Laravel MVC Architecture
 
Pigging Solutions in Pet Food Manufacturing
Pigging Solutions in Pet Food ManufacturingPigging Solutions in Pet Food Manufacturing
Pigging Solutions in Pet Food Manufacturing
 
08448380779 Call Girls In Civil Lines Women Seeking Men
08448380779 Call Girls In Civil Lines Women Seeking Men08448380779 Call Girls In Civil Lines Women Seeking Men
08448380779 Call Girls In Civil Lines Women Seeking Men
 
The 7 Things I Know About Cyber Security After 25 Years | April 2024
The 7 Things I Know About Cyber Security After 25 Years | April 2024The 7 Things I Know About Cyber Security After 25 Years | April 2024
The 7 Things I Know About Cyber Security After 25 Years | April 2024
 
From Event to Action: Accelerate Your Decision Making with Real-Time Automation
From Event to Action: Accelerate Your Decision Making with Real-Time AutomationFrom Event to Action: Accelerate Your Decision Making with Real-Time Automation
From Event to Action: Accelerate Your Decision Making with Real-Time Automation
 
Install Stable Diffusion in windows machine
Install Stable Diffusion in windows machineInstall Stable Diffusion in windows machine
Install Stable Diffusion in windows machine
 
My Hashitalk Indonesia April 2024 Presentation
My Hashitalk Indonesia April 2024 PresentationMy Hashitalk Indonesia April 2024 Presentation
My Hashitalk Indonesia April 2024 Presentation
 
Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...
Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...
Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...
 
Integration and Automation in Practice: CI/CD in Mule Integration and Automat...
Integration and Automation in Practice: CI/CD in Mule Integration and Automat...Integration and Automation in Practice: CI/CD in Mule Integration and Automat...
Integration and Automation in Practice: CI/CD in Mule Integration and Automat...
 
Presentation on how to chat with PDF using ChatGPT code interpreter
Presentation on how to chat with PDF using ChatGPT code interpreterPresentation on how to chat with PDF using ChatGPT code interpreter
Presentation on how to chat with PDF using ChatGPT code interpreter
 
WhatsApp 9892124323 ✓Call Girls In Kalyan ( Mumbai ) secure service
WhatsApp 9892124323 ✓Call Girls In Kalyan ( Mumbai ) secure serviceWhatsApp 9892124323 ✓Call Girls In Kalyan ( Mumbai ) secure service
WhatsApp 9892124323 ✓Call Girls In Kalyan ( Mumbai ) secure service
 
Factors to Consider When Choosing Accounts Payable Services Providers.pptx
Factors to Consider When Choosing Accounts Payable Services Providers.pptxFactors to Consider When Choosing Accounts Payable Services Providers.pptx
Factors to Consider When Choosing Accounts Payable Services Providers.pptx
 
04-2024-HHUG-Sales-and-Marketing-Alignment.pptx
04-2024-HHUG-Sales-and-Marketing-Alignment.pptx04-2024-HHUG-Sales-and-Marketing-Alignment.pptx
04-2024-HHUG-Sales-and-Marketing-Alignment.pptx
 
08448380779 Call Girls In Greater Kailash - I Women Seeking Men
08448380779 Call Girls In Greater Kailash - I Women Seeking Men08448380779 Call Girls In Greater Kailash - I Women Seeking Men
08448380779 Call Girls In Greater Kailash - I Women Seeking Men
 
FULL ENJOY 🔝 8264348440 🔝 Call Girls in Diplomatic Enclave | Delhi
FULL ENJOY 🔝 8264348440 🔝 Call Girls in Diplomatic Enclave | DelhiFULL ENJOY 🔝 8264348440 🔝 Call Girls in Diplomatic Enclave | Delhi
FULL ENJOY 🔝 8264348440 🔝 Call Girls in Diplomatic Enclave | Delhi
 
Pigging Solutions Piggable Sweeping Elbows
Pigging Solutions Piggable Sweeping ElbowsPigging Solutions Piggable Sweeping Elbows
Pigging Solutions Piggable Sweeping Elbows
 
Key Features Of Token Development (1).pptx
Key  Features Of Token  Development (1).pptxKey  Features Of Token  Development (1).pptx
Key Features Of Token Development (1).pptx
 
How to Remove Document Management Hurdles with X-Docs?
How to Remove Document Management Hurdles with X-Docs?How to Remove Document Management Hurdles with X-Docs?
How to Remove Document Management Hurdles with X-Docs?
 
How to convert PDF to text with Nanonets
How to convert PDF to text with NanonetsHow to convert PDF to text with Nanonets
How to convert PDF to text with Nanonets
 

Two Algorithms Weakly Supervised Denoising EEG Data

  • 1. Two Algorithms for Weakly Supervised Denoising of EEG Data Tim Oates Professor University of Maryland Baltimore County
  • 2. The Real Brains 3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 2 Sunil Gandhi Ph.D. student Will graduate soon!
  • 3. Agenda • Introduction To Artifact Removal From EEG data • ICA And Multi-instance Learning Solution • Asymmetric Generative Adversarial Network Solution • Future Work • Questions 3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 3
  • 4. • Introduction To Artifact Removal From EEG data • ICA And Multi-instance Learning Solution • Asymmetric Generative Adversarial Network Solution • Future Work • Questions 3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 4
  • 5. Electroencephalography (EEG) • Electroencephalography (EEG) is used to record electrical activity in the brain 5 Image Source: https://en.wikipedia.org/wiki/Electroencephalography 3/2/2019 Artifact Removal From Electroencephalography (EEG) Data
  • 6. Electroencephalography (EEG) Data • Electroencephalography (EEG) is used to record electrical activity in the brain – Seizure detection 6 Page, Adam, Chris Sagedy, Emily Smith, Nasrin Attaran, Tim Oates, and Tinoosh Mohsenin. "A flexible multichannel EEG feature extractor and classifier for seizure detection." IEEE Transactions on Circuits and Systems II: Express Briefs 62, no. 2 (2015): 109-113. 3/2/2019 Artifact Removal From Electroencephalography (EEG) Data
  • 7. Electroencephalography (EEG) Data • Electroencephalography (EEG) is used to record electrical activity in the brain – Seizure detection – Determining cognitive load 7 Bashivan, Pouya, Irina Rish, Mohammed Yeasin, and Noel Codella. "Learning representations from EEG with deep recurrent- convolutional neural networks." International Conference on Learning Representations 2016. 3/2/2019 Artifact Removal From Electroencephalography (EEG) Data
  • 8. Electroencephalography (EEG) Data • Electroencephalography (EEG) is used to record electrical activity in the brain – Seizure detection – Determining cognitive load – Brain Computer interfaces (BCI) 8 Source: https://www.youtube.com/watch?v=rYwgnFeFqmc 3/2/2019 Artifact Removal From Electroencephalography (EEG) Data
  • 9. Electroencephalography (EEG) Data • Electroencephalography (EEG) is used to record electrical activity in the brain – Seizure detection – Determining cognitive load – Brain Computer interfaces (BCI) – Driver drowsiness estimation 9 Lance, Brent J., W. David Hairston, Greg Apker, Keith W. Whitaker, Geoff Slipher, Randy Mrozek, Scott E. Kerick, Jason Metcalfe, Christopher Manteuffel, and Matthew Jaswa. 2012 year-end report on neurotechnologies for in-vehicle applications. No. ARL-SR-267. ARMY RESEARCH LAB ABERDEEN PROVING GROUND MD, 2013. 3/2/2019 Artifact Removal From Electroencephalography (EEG) Data
  • 10. Challenges in using EEG data • Compact comfortable headsets • Noisy Data • Online, fully automated algorithms • Hardware implementations 3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 10
  • 11. Challenges in using EEG data • Compact comfortable headsets • Noisy Data • Online, fully automated algorithms • Hardware implementations 3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 11
  • 12. Challenges in using EEG data • Compact comfortable headsets • Noisy Data • Online, fully automated algorithms • Hardware implementations 3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 12
  • 13. EEG Artifacts • Artifacts: Unwanted electrical activity arising from sources other than the brain • Types of Artifacts – Biological Artifacts • Ocular artifacts • Muscular artifacts • Cardiac artifacts – External artifacts • Electrode motion • External device artifacts 3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 13
  • 14. EEG Artifacts • Artifacts: Unwanted electrical activity arising from sources other than the brain • Types of Artifacts – Biological Artifacts • Ocular artifacts • Muscular artifacts • Cardiac artifacts – External artifacts • Electrode motion • External device artifacts 3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 14 Eye Blink Artifact
  • 15. EEG Artifacts • Artifacts: Unwanted electrical activity arising from sources other than the brain • Types of Artifacts – Biological Artifacts – External Artifacts • Why is it important to remove them? – Increase the chance of false alarms in seizure detection (Seneviratne et al. 2013) – They can also alter the shape of neurological events causing unintentional control of BCI systems (Vaughan et al. 2003). 3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 15
  • 16. EEG Artifacts • Artifacts: Unwanted electrical activity arising from sources other than the brain • Types of Artifacts – Biological Artifacts – External Artifacts • Why is it important to remove them? • Can occurrence of artifacts be avoided? – The occurrence of external artifacts can be reduced by proper placement of electrodes, but it is impossible to avoid artifacts of biological origin. 3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 16
  • 17. Contributions • Propose a system for artifact removal from EEG data using weak supervisory information. • Improve the model by proposing an online, fully automated, end-to-end system for artifact removal trained using unpaired training corpora. • Creating a framework for evaluation of artifact removal algorithms • Improving artifact removal using additional annotations • Generalizing AsymmetricGAN for denoising speech data 3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 17
  • 18. Contributions • Propose a system for artifact removal from EEG data using weak supervisory information. • Improve the model by proposing an online, fully automated, end-to-end system for artifact removal trained using unpaired training corpora. • Creating a framework for evaluation of artifact removal algorithms • Improving artifact removal using additional annotations • Generalizing AsymmetricGAN for denoising speech data 3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 18
  • 19. • Introduction To Artifact Removal From EEG data • ICA And Multi-instance Learning Solution • Asymmetric Generative Adversarial Network Solution • Future Work • Questions 3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 19
  • 20. Related Work • Artifact Rejection(Nolan, Whelan, and Reilly 2010) – Train an artifact detector – Remove the segments of data where artifacts are present – Excessive loss of information • Regression using reference channel (Croft and Barry 2000) – Use a reference channel like Electrocardiography (ECG), Electrooculography (EOG) or Electromyography (EMG). – Fail if a reference signal is not available. • Adaptive filtering (Sweeney, Ward, and McLoone 2012) – Generate an artifact signal that is uncorrelated to the EEG signal but correlated to the reference signal. – Fail if a reference signal is not available. 3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 20
  • 21. Related Work 3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 21 • Wavelet transforms (Islam et al. 2017) (Daly et al. 2015) – Decompose each channel using DWT – Threshold the low frequency component and reconstruct the signal – Process each channel separately and could miss important clues for removing the artifact.
  • 22. 22
  • 23. Related Work • Principal component analysis (PCA) (Berg and Scherg 1991) – One of the initial approaches that used principal component analysis (PCA) for decomposing the signal and removing artifacts from the decomposed signal. – Uses orthogonal transformation to convert a multi channel EEG signal into a set of linearly uncorrelated variables. – It has been demonstrated that PCA is unable to separate some artifactual components from brain signals, especially when they have similar amplitudes (Fitzgibbon et al. 2007b). 3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 23
  • 24. Independent component analysis (ICA) • Most popular technique for artifact removal from EEG data • ICA is used to recover independent source signals called components and then the components corresponding to artifacts are identified. 3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 24
  • 25. ICA Based Artifact Removal • Shackman, Alexander J., et al. "Identifying robust and sensitive frequency bands for interrogating neural oscillations." Neuroimage 51.4 (2010): 1319-1333. • Fitzgibbon, S. P., et al. "Removal of EEG noise and artifact using blind source separation." Journal of Clinical Neurophysiology 24.3 (2007): 232-243. • Joyce, Carrie A., Irina F. Gorodnitsky, and Marta Kutas. "Automatic removal of eye movement and blink artifacts from EEG data using blind component separation." Psychophysiology 41.2 (2004): 313-325. • Jung, Tzyy-Ping, et al. "Removal of eye activity artifacts from visual event-related potentials in normal and clinical subjects." Clinical Neurophysiology 111.10 (2000): 1745-1758. • Jung, Tzyy-Ping, et al. "Removing electroencephalographic artifacts by blind source separation." Psychophysiology 37.02 (2000): 163-178. • Keren, Alon S., Shlomit Yuval-Greenberg, and Leon Y. Deouell. "Saccadic spike potentials in gamma-band EEG: characterization, detection and suppression." Neuroimage 49.3 (2010): 2248-2263. 253/2/2019 Artifact Removal From Electroencephalography (EEG) Data
  • 26. ICA Based Artifact Removal • Shackman, Alexander J., et al. "Identifying robust and sensitive frequency bands for interrogating neural oscillations." Neuroimage 51.4 (2010): 1319-1333. • Fitzgibbon, S. P., et al. "Removal of EEG noise and artifact using blind source separation." Journal of Clinical Neurophysiology 24.3 (2007): 232-243. • Joyce, Carrie A., Irina F. Gorodnitsky, and Marta Kutas. "Automatic removal of eye movement and blink artifacts from EEG data using blind component separation." Psychophysiology 41.2 (2004): 313-325. • Jung, Tzyy-Ping, et al. "Removal of eye activity artifacts from visual event-related potentials in normal and clinical subjects." Clinical Neurophysiology 111.10 (2000): 1745-1758. • Jung, Tzyy-Ping, et al. "Removing electroencephalographic artifacts by blind source separation." Psychophysiology 37.02 (2000): 163-178. • Keren, Alon S., Shlomit Yuval-Greenberg, and Leon Y. Deouell. "Saccadic spike potentials in gamma-band EEG: characterization, detection and suppression." Neuroimage 49.3 (2010): 2248-2263. Artifact components need to be manually identified or supervisory signal is needed for training 263/2/2019 Artifact Removal From Electroencephalography (EEG) Data
  • 27. ICA Based Artifact Removal • Shackman, Alexander J., et al. "Identifying robust and sensitive frequency bands for interrogating neural oscillations." Neuroimage 51.4 (2010): 1319-1333. • Fitzgibbon, S. P., et al. "Removal of EEG noise and artifact using blind source separation." Journal of Clinical Neurophysiology 24.3 (2007): 232-243. • Joyce, Carrie A., Irina F. Gorodnitsky, and Marta Kutas. "Automatic removal of eye movement and blink artifacts from EEG data using blind component separation." Psychophysiology 41.2 (2004): 313-325. • Jung, Tzyy-Ping, et al. "Removal of eye activity artifacts from visual event-related potentials in normal and clinical subjects." Clinical Neurophysiology 111.10 (2000): 1745-1758. • Jung, Tzyy-Ping, et al. "Removing electroencephalographic artifacts by blind source separation." Psychophysiology 37.02 (2000): 163-178. • Keren, Alon S., Shlomit Yuval-Greenberg, and Leon Y. Deouell. "Saccadic spike potentials in gamma-band EEG: characterization, detection and suppression." Neuroimage 49.3 (2010): 2248-2263. Key Idea: Identify components corresponding to artifacts using weak supervisory signal 273/2/2019 Artifact Removal From Electroencephalography (EEG) Data
  • 28. System Architecture 28 Mul$ Channel EEG Signal Channel 1 Channel 2 Channel 64 . ... .. ICA Bag of components . .. SAX Feature Extrac0on . .. 0 1 0 1 0 0 Mul0- Instance Learning (MIL) 0 0 1 1 0 0 0 0 1 1 1 1 Vector represen$ng component Sax string: "aaa" "abc" "aab" "aba" "acb" "acc" Probability of noise for each component Ar$fact/ Not Ar$fact? • Main Processing Blocks ⎻ Independent Component Analysis ⎻ SAX feature extraction ⎻ Multi Instance Learning 3/2/2019 Artifact Removal From Electroencephalography (EEG) Data
  • 29. Independent component analysis (ICA) 3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 29 ICA Origin 1 Origin 2 Origin 3 Mixing System Observed 1 Observed 2 Observed 3 Estimated 1 Estimated 2 Estimated 3 Unknown
  • 30. Independent component analysis (ICA) 3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 30 Source: http://scikit-learn.sourceforge.net/0.6/auto_examples/plot_ica_blind_source_separation.html
  • 31. Independent component analysis (ICA) 3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 31 ICA Origin 1 Origin 2 Origin 3 Mixing System Observed 1 Observed 2 Observed 3 Estimated 1 Estimated 2 Estimated 3 Unknown
  • 32. Independent component analysis (ICA) 3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 32 ICA Clean EEG Signal Artifact Mixing System Observed 1 Observed 2 Observed 3 Denoised EEG Artifact Unknown
  • 33. System Architecture 33 Mul$ Channel EEG Signal Channel 1 Channel 2 Channel 64 . ... .. ICA Bag of components . .. SAX Feature Extrac0on . .. 0 1 0 1 0 0 Mul0- Instance Learning (MIL) 0 0 1 1 0 0 0 0 1 1 1 1 Vector represen$ng component Sax string: "aaa" "abc" "aab" "aba" "acb" "acc" Probability of noise for each component Ar$fact/ Not Ar$fact? • Main Processing Blocks ⎻ Independent Component Analysis ⎻ SAX feature extraction ⎻ Multi Instance Learning 3/2/2019 Artifact Removal From Electroencephalography (EEG) Data
  • 34. Slide by Eamonn Keogh(http://www.cs.ucr.edu/~eamonn/SAX.ppt) 34 How do we obtain SAX features? 0 00 20 40 60 80 100 120 bb b a c c c a baabccbc 0 20 20 40 60 80 100 120 3/2/2019 Artifact Removal From Electroencephalography (EEG) Data
  • 35. System Architecture 35 Mul$ Channel EEG Signal Channel 1 Channel 2 Channel 64 . ... .. ICA Bag of components . .. SAX Feature Extrac0on . .. 0 1 0 1 0 0 Mul0- Instance Learning (MIL) 0 0 1 1 0 0 0 0 1 1 1 1 Vector represen$ng component Sax string: "aaa" "abc" "aab" "aba" "acb" "acc" Probability of noise for each component Ar$fact/ Not Ar$fact? • Main Processing Blocks ⎻ Independent Component Analysis ⎻ SAX feature extraction ⎻ Multi Instance Learning 3/2/2019 Artifact Removal From Electroencephalography (EEG) Data
  • 37. Multi-Instance Learning • A bag is labelled positive if it has at least one positive instance • Goal: Predict bag label and instance labels Bag Positive/Negative Bag? 373/2/2019 Artifact Removal From Electroencephalography (EEG) Data
  • 38. Multi-Instance Learning • A bag is labelled positive if it has at least one positive instance • Goal: Predict bag label and instance label Bag Positive/Negative Instance? Positive/Negative Instance? Positive/Negative Instance? Positive/Negative Instance? 383/2/2019 Artifact Removal From Electroencephalography (EEG) Data
  • 39. System Architecture 39 Mul$ Channel EEG Signal Channel 1 Channel 2 Channel 64 . ... .. ICA Bag of components . .. SAX Feature Extrac0on . .. 0 1 0 1 0 0 Mul0- Instance Learning (MIL) 0 0 1 1 0 0 0 0 1 1 1 1 Vector represen$ng component Sax string: "aaa" "abc" "aab" "aba" "acb" "acc" Probability of noise for each component Ar$fact/ Not Ar$fact? • Main Processing Blocks ⎻ Independent Component Analysis ⎻ SAX feature extraction ⎻ Multi Instance Learning 3/2/2019 Artifact Removal From Electroencephalography (EEG) Data
  • 40. EEG Dataset Generated by the US Army Research Laboratory 40 EEG when participant is not performing any activity • 17 subjects • Artifacts o Sync pulse o Clench jaw o Move jaw vertically o Blink (no squinting!) o Move eyes left, then back to center o Move eyes up, then back to center o Raise and lower eyebrows o Rotate head side-to-side o Shrug shoulders o Rotate torso (hips) EEG when participant is raising and lowering eyebrows 3/2/2019 Artifact Removal From Electroencephalography (EEG) Data
  • 41. Results 41 Algorithm Kernel Accuracy Normalized Set Kernel Linear 95.2% MISVM Linear 73.2% miSVM Linear 70.0% MISVM Quadratic Kernel 81.6% miSVM Quadratic Kernel 66.0% 3/2/2019 Artifact Removal From Electroencephalography (EEG) Data
  • 42. Results 3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 42 Patient ID LOSO Accuracy Personalized Accuracy 1 72.54901961 98.04(3.39) 2 76.47058824 96.08(3.39) 3 76.47058824 92.16(6.79) 4 82.35294118 98.04(3.39) 5 84.31372549 98.04(3.39) 6 31.37254902 94.12(0) 7 94.11764706 100(0) 8 100 100(0) 9 98.03921569 100(0) 10 66.66666667 90.2(6.8) 11 98.03921569 100(0) 12 80.39215686 94.12(10.19) 13 88.23529412 100(0) 14 49.01960784 100(0) 15 66.66666667 90.2(12.25) 16 54.90196078 100(0) 17 90.19607843 100(0) Average 77.0472895 97.12
  • 43. Accuracy 3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 43 Patient ID LOSO Accuracy Personalized Accuracy 1 72.54901961 98.04(3.39) 2 76.47058824 96.08(3.39) 3 76.47058824 92.16(6.79) 4 82.35294118 98.04(3.39) 5 84.31372549 98.04(3.39) 6 31.37254902 94.12(0) 7 94.11764706 100(0) 8 100 100(0) 9 98.03921569 100(0) 10 66.66666667 90.2(6.8) 11 98.03921569 100(0) 12 80.39215686 94.12(10.19) 13 88.23529412 100(0) 14 49.01960784 100(0) 15 66.66666667 90.2(12.25) 16 54.90196078 100(0) 17 90.19607843 100(0) Average 77.0472895 97.12 Patient ID Accuracy 1 100(0) 2 89(7.3) 3 100(0) 4 99(2.8) 5 100((0) 6 97(5.5) 7 97(5.5) Average 98(5.5) Pooled 97(1.9) Bootstrap 14(17.7) LOSO 75 Lawhern, Vernon, W. David Hairston, Kaleb McDowell, Marissa Westerfield, and Kay Robbins. "Detection and classification of subject-generated artifacts in EEG signals using autoregressive models." Journal of neuroscience methods 208, no. 2 (2012): 181-189.
  • 44. Summary • We presented an EEG artifact removal system, using ICA, SAX feature extraction, and Multi-Instance Learning algorithms. The proposed system uses a weak supervisory signal to indicate that some noise is occurring, but not what the source of the noise is or how it is manifested in the EEG signal. We optimize the hyper-parameters of the system to reduce the execution time of the system while maintaining accuracy. 45 Jafari, Ali, Sunil Gandhi, Sri Harsha Konuru, W. David Hairston, Tim Oates, and Tinoosh Mohsenin. "An EEG artifact identification embedded system using ICA and multi-instance learning." In Circuits and Systems (ISCAS), 2017 IEEE International Symposium on, pp. 1-4. IEEE, 2017. 3/2/2019 Artifact Removal From Electroencephalography (EEG) Data
  • 45. Limitations • Independent component analysis needs large number of samples to converge. ICA has high computational complexity and large memory requirements, making it unsuitable for real- time applications. • Each subsystem has its own hyperparameters and tuning them jointly is a challenging task. 463/2/2019 Artifact Removal From Electroencephalography (EEG) Data
  • 46. Limitations • Independent component analysis needs large number of samples to converge. ICA has high computational complexity and large memory requirements, making it unsuitable for real- time applications. • Each subsystem has its own hyperparameters and tuning them jointly is a challenging task. Also, 47 Can we learn an end to end system to generate clean EEG from noisy EEG using deep neural networks 3/2/2019 Artifact Removal From Electroencephalography (EEG) Data
  • 47. • Introduction To Artifact Removal From EEG data • ICA And Multi-instance Learning Solution • Asymmetric Generative Adversarial Network Solution • Future Work • Questions 3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 48
  • 48. Background • Generative Adversarial network (GAN) • CycleGAN: Unpaired Image-to-Image Translation using Cycle- Consistent Adversarial Networks 3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 49
  • 49. Generative Adversarial Networks • Generator network: try to fool the discriminator by generating real-looking images • Discriminator network: try to distinguish between real and fake images real or fake? Discriminator z G(z) D Generator G 3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 50
  • 50. [Goodfellow et al. 2014] real z G(z) DG D x fake 51
  • 51. CycleGAN: Unpaired Image-to-image translation using cycle-consistent adversarial networks 52
  • 52. … [Zhu*, Park*, Isola, Efros on arxiv] …… X Dx Y Dy G F Cycle GAN 53
  • 53. Cycle-consistency Loss Backward cycle loss: 𝐺 𝐹 𝑦 − 𝑦 1 Forward cycle loss: F G x − x 1 G(x) F(G x )x F(y) G(F x )𝑦 54
  • 54. CycleGAN: Unpaired Image-to-image translation using cycle-consistent adversarial networks EEG Data EEG Data without Artifacts 55
  • 55. CycleGAN Model A B G_B G_A Da Db 3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 56
  • 56. CycleGAN Model A B G_B G_A Da Db • Problems – Network G_A does not have access to predicted noise signal to reconstruct A – The network is penalized for not adding noise that is the same as predicted noise 3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 57
  • 57. CycleGAN Model A B G_B G_A Da Db A A’ G_B G_A 3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 58
  • 58. AsymmetricGAN Model A B G_B G_A Da Db We preserve the noise in the signal and use it for reconstruction of the original signal 3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 59 A BN G_B G_NG_N + Da Db
  • 59. AsymmetricGAN Model A B G_B G_A Da Db + 3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 60
  • 60. AsymmetricGAN Model 3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 61 G_N is a function that extracts noise if the input is a noisy signal A. It generates noise if the input is a clean signal B.
  • 61. Generator and discriminator network architecture 3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 62
  • 62. Generator and discriminator network architecture 3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 63 • Network is independent of the relative ordering of the channels • Channel indices do not correspond to the spatial locations of the electrodes. • Filter Size: • 1D convolution with filter size K • 2D convolution with filter size N ✕ k Qi, Charles R., Hao Su, Kaichun Mo, and Leonidas J. Guibas. "Pointnet: Deep learning on point sets for 3d classification and segmentation." Proc. Computer Vision and Pattern Recognition (CVPR), IEEE 1, no. 2 (2017): 4.
  • 63. Evaluation • Evaluation of artifact removal from EEG data is difficult as ground truth signal does not exist. • We just have set of clean EEG signals and noisy EEG signals. We don’t have clean version of noisy EEG. • Visualizing and understanding EEG data is a time consuming task. • To validate the model we create a simpler synthetic dataset. 3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 64
  • 64. Synthetic Dataset • Clean signal is a linear combination of a sine and a square wave • Noisy signal is linear combination of sine, square and sawtooth waves • The period of the sine and square waves is randomly selected between 2 and 5. Sawtooth has fixed period of 6 • Number of samples in each time series is 1000 • Training set: 4000 Signals • Validation set: 1000 Signals • Test set: 100 Signals 3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 65
  • 65. Synthetic Dataset 3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 66 (e) True Sources (c) Signal cleaned by Asymmetric GAN (a) Signal with Artifact (b) Clean Signal (d) Noise Signal predicted by Asymmetric GAN
  • 66. Synthetic Dataset 3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 67 (e) True Sources (c) Signal cleaned by Asymmetric GAN (a) Signal with Artifact (b) Clean Signal (d) Noise Signal predicted by Asymmetric GAN The MSE between the ground truth clean signal and denoised signal is 0.0406. The mean MSE error for entire test set is 0.0387 and standard deviation is 0.0043.
  • 67. EEG Dataset Generated by the US Army Research Laboratory • During collection of this dataset, participants were told to not move and look straight at the computer screen for the collection of clean data. • Despite the instruction to the patients, we noticed that there were artifacts even in “clean” data. We also noticed that some “noisy” EEG did not contain the corresponding artifact. • Manually annotating all artifacts from all channels for all artifacts is a time consuming task. So in this work, we focus on ocular artifacts in the Fp1 electrode of the frontal region. • Remove all patients that have more than two ocular artifacts in the clean data and do not have artifacts in the region of eyebrow raising. • In the resulting dataset, we have 4 patients with clean data and 10 patients with noisy data. • Each patient’s clean data has 4836 samples and noisy data has 420354 samples. We use this manually annotated data in all experiments below. 3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 68
  • 68. Evaluation • Qualitative Results • Using a detector 3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 69
  • 69. Qualitative Results (a) (b) 3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 70
  • 70. Evaluation by detection 3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 71 • We use artifact detection as a way of measuring the performance of artifact removal. • We first train artifact detection and artifact removal and use the artifact detector to classify every window in EEG data denoised by the artifact removal algorithm. • The error is given by the percent of windows where an artifact was detected in denoised EEG data.
  • 71. Dataset Division 3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 72 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 Clean Noisy Clean Noisy Clean Noisy Clean Noisy Train Test Train Test Artifact Detector Artifact Removal
  • 72. Evaluation by detection • Artifact Detector accuracy: 97.39% • Denoised signal was classified as clean by artifact detector 72.37% times. 3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 73
  • 73. Summary • We presented an online, fully automated, end-to-end system for denoising EEG data. Our system for denoising is trained using unpaired training corpora. It does not need any information about the source of the noise or how it is manifested in the EEG signal. We created a synthetic dataset and used it to validate our network. We also used our system to remove artifacts from existing EEG dataset 3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 74 Sunil Gandhi, Tim Oates, Tinoosh Mohsenin, Dave Hairston. ”Denoising Time Series Data Using Asymmetric Generative Adversarial Networks.” In Pacific-Asia Conference on Knowledge Discovery and Data Mining, 2018.
  • 74. • Introduction To Artifact Removal From EEG data • ICA And Multi-instance Learning Solution • Asymmetric Generative Adversarial Network Solution • Future Work • Questions 3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 75
  • 75. Future Work • Creating a framework for evaluation of artifact removal algorithms • Removing multiple artifacts using AsymmetricGAN • Improving artifact removal using additional annotations • Generalizing AsymmetricGAN for speech data 3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 76
  • 76. Evaluation of artifact removal algorithms • Evaluating artifact removal algorithms is important before they can be used in clinical contexts. • But, it is challenging because of a lack of availability of ground truth clean EEG signals. 3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 77
  • 77. Evaluation of artifact removal algorithms • Evaluating artifact removal algorithms is important before they can be used in clinical contexts. • But, it is challenging because of a lack of availability of ground truth clean EEG signals. 3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 78 The performance evaluation of artifact removal methods found in the literature is always problematic. It can be done either by visually by expert which is subjective or by synthetic/semi-synthetic data (but uncertainty of reconstructed data whether perfectly realistic or not). Since there is neither any ground truth data available nor any universal or standard quantitative metric(s) used in the literature that can capture both amount of artifact removal and distortion. … Therefore, it is not fair to tell which performs best based on the study. Islam, Md Kafiul, Amir Rastegarnia, and Zhi Yang. "Methods for artifact detection and removal from scalp EEG: A review." Neurophysiologie Clinique/Clinical Neurophysiology 46, no. 4-5 (2016): 287-305.
  • 78. Evaluation of artifact removal algorithms • Qualitative Evaluation • Evaluation using simulated data • Evaluation using detection • Correlating with the reference signal 3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 79
  • 79. Evaluation of artifact removal algorithms • Qualitative Evaluation – Subjective – Does not give quantitative metric 3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 80
  • 80. Evaluation of artifact removal algorithms • Qualitative Evaluation • Evaluation using simulated data – Allows usage of metrics like signal-to-noise ratio (SNR) and normalized mean squared error to compare deviation between the denoised signal and clean signal. – Training and evaluation can be performed on large datasets – But it is very difficult to replicate all the characteristics of EEG data like synchronization among channels, time-locking to ERPs and contamination by different types of artifacts in a realistic manner. 3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 81
  • 81. Evaluation of artifact removal algorithms • Qualitative Evaluation • Evaluation using simulated data • Evaluation using detection – Use artifact detector to check if artifacts are present in the denoised signal. – Can be used to evaluate denoising of real EEG data – Automated, not subjective and gives a quantitative measure to compare several artifact removal algorithms. – Usefulness of this metric is dependent on the performance of the artifact detector. – Does not give a measure of the sensitivity (whether the method removes the artifacts) and specificity (whether it preserves EEG signals) of the artifact removal algorithm. 3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 82
  • 82. Evaluation of artifact removal algorithms • Qualitative Evaluation • Evaluation using simulated data • Evaluation using detection • Correlating with the reference signal – Cannot be performed if a reference signal for the corresponding artifact is not available. – Does not give a measure of the sensitivity (whether the method removes the artifacts) and specificity (whether it preserves EEG signals) of the artifact removal algorithm. 3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 83
  • 83. Evaluation of artifact removal algorithms • Evaluation Methods – Qualitative Evaluation – Evaluation using simulated data – Evaluation using detection – Correlating with the reference signal • Multiple evaluation methods with several variants of each of these methods exist in literature. • Each of these methods has their strengths and limitations. This makes performance comparison of these methods hard. • In future we plan to – Standardize the evaluation mechanism – Create a tool for evaluating artifact removal algorithms using all of the above approaches. – Comparing our approach with existing methods using the tool 3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 84
  • 84. Future Work • Creating a framework for evaluation of artifact removal algorithms • Removing multiple artifacts using AsymmetricGAN • Improving artifact removal using additional annotations • Generalizing AsymmetricGAN for speech data 3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 85
  • 85. Removing multiple artifacts using AsymmetricGAN • Having separate algorithms for removal of different types of artifacts makes the preprocessing step inefficient. • In future, we plan to adapt the AsymmetricGAN architecture for removing multiple types of artifacts. 3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 86 G_B G_NG_N + Da Db MAA B OA
  • 86. Future Work • Creating a framework for evaluation of artifact removal algorithms • Removing multiple artifacts using AsymmetricGAN • Improving artifact removal using additional annotations • Generalizing AsymmetricGAN for speech data 3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 87
  • 87. Improving artifact removal using additional annotations • Modify AsymmetricGAN to utilize – artifact type – location of artifact – reference channel. • We will study differences in performance because of each type of additional data. • This will give insight on which annotations are most helpful for artifact removal algorithms, thus helping creation of future datasets. 3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 88
  • 88. Future Work • Creating a framework for evaluation of artifact removal algorithms • Removing multiple artifacts using AsymmetricGAN • Improving artifact removal using additional annotations • Generalizing AsymmetricGAN for speech data 3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 89
  • 89. Generalizing AsymmetricGAN for speech data • Speech denoising produces noise-free speech signals from noisy recordings • Applications – Recognizing speech in environments with background noise like car driving. – Reduce discomfort and increase understanding of the speech for people wearing hearing aids • Easier to evaluate • Evaluate generalizability of asymmetricGAN across domains • Datasets: – Voice Bank corpus: Speech data for 400 sentences from 28 speakers each. – Demand dataset: Real world background noise of 18 diverse environments 3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 90
  • 90. Questions? 3/2/2019 Artifact Removal From Electroencephalography (EEG) Data 91

Editor's Notes

  1. Technique for monitoring brain activity
  2. Wu, Dongrui, Vernon J. Lawhern, Stephen Gordon, Brent J. Lance, and Chin-Teng Lin. "Offline EEG-based driver drowsiness estimation using enhanced batch-mode active learning (EBMAL) for regression." In Systems, Man, and Cybernetics (SMC), 2016 IEEE International Conference on, pp. 000730-000736. IEEE, 2016.Harvard Lance, Brent J., W. David Hairston, Greg Apker, Keith W. Whitaker, Geoff Slipher, Randy Mrozek, Scott E. Kerick, Jason Metcalfe, Christopher Manteuffel, and Matthew Jaswa. 2012 year-end report on neurotechnologies for in-vehicle applications. No. ARL-SR-267. ARMY RESEARCH LAB ABERDEEN PROVING GROUND MD, 2013. Offline EEG-Based Driver Drowsiness Estimation Using Enhanced Batch-Mode Active Learning (EBMAL) for Regression assess the state of human vehicle operators and integrate that state information into future vehicle system
  3. Wu, Dongrui, Vernon J. Lawhern, Stephen Gordon, Brent J. Lance, and Chin-Teng Lin. "Offline EEG-based driver drowsiness estimation using enhanced batch-mode active learning (EBMAL) for regression." In Systems, Man, and Cybernetics (SMC), 2016 IEEE International Conference on, pp. 000730-000736. IEEE, 2016.Harvard Lance, Brent J., W. David Hairston, Greg Apker, Keith W. Whitaker, Geoff Slipher, Randy Mrozek, Scott E. Kerick, Jason Metcalfe, Christopher Manteuffel, and Matthew Jaswa. 2012 year-end report on neurotechnologies for in-vehicle applications. No. ARL-SR-267. ARMY RESEARCH LAB ABERDEEN PROVING GROUND MD, 2013. Offline EEG-Based Driver Drowsiness Estimation Using Enhanced Batch-Mode Active Learning (EBMAL) for Regression assess the state of human vehicle operators and integrate that state information into future vehicle system
  4. Page, Adam, Chris Sagedy, Emily Smith, Nasrin Attaran, Tim Oates, and Tinoosh Mohsenin. "A flexible multichannel EEG feature extractor and classifier for seizure detection." IEEE Transactions on Circuits and Systems II: Express Briefs 62, no. 2 (2015): 109-113. Wu, Dongrui, Vernon J. Lawhern, Stephen Gordon, Brent J. Lance, and Chin-Teng Lin. "Offline EEG-based driver drowsiness estimation using enhanced batch-mode active learning (EBMAL) for regression." In Systems, Man, and Cybernetics (SMC), 2016 IEEE International Conference on, pp. 000730-000736. IEEE, 2016.Harvard Lance, Brent J., W. David Hairston, Greg Apker, Keith W. Whitaker, Geoff Slipher, Randy Mrozek, Scott E. Kerick, Jason Metcalfe, Christopher Manteuffel, and Matthew Jaswa. 2012 year-end report on neurotechnologies for in-vehicle applications. No. ARL-SR-267. ARMY RESEARCH LAB ABERDEEN PROVING GROUND MD, 2013. Offline EEG-Based Driver Drowsiness Estimation Using Enhanced Batch-Mode Active Learning (EBMAL) for Regression assess the state of human vehicle operators and integrate that state information into future vehicle system
  5. Wu, Dongrui, Vernon J. Lawhern, Stephen Gordon, Brent J. Lance, and Chin-Teng Lin. "Offline EEG-based driver drowsiness estimation using enhanced batch-mode active learning (EBMAL) for regression." In Systems, Man, and Cybernetics (SMC), 2016 IEEE International Conference on, pp. 000730-000736. IEEE, 2016.Harvard assess the state of human vehicle operators and integrate that state information into future vehicle system There are bunch of other applications that I won’t go into.
  6. Some electrodes
  7. There is a strong need for method for denoising the EEG data
  8. We saw few slides back that it is important problem and large number of methods have been proposed
  9. Users with relatively short data files, for whom segment rejection would lead to an unacceptably low remaining number of segments for analysis, may choose an optional post-segmentation step involving the interpolation of data for channels determined to be artifact-contaminated within each individual segment, as implemented by FASTER software (Nolanet al., 2010). Each channel in each segment is evaluated onthe four FASTER criteria (variance, median gradient, amplitude range, and deviation from mean amplitude), and the Z score(a measure of standard deviation from the mean) for each channel in that segment is generated for each of the fourmetrics. Any channels with one or more Z scores that are greater than 3 standard deviations from the mean for an individual segment are marked bad for that segment. These criteria may identify segments with residual high-amplitude artifacts(e.g., eye artifacts), electrode discontinuity (e.g., electrode has lost contact with the scalp temporarily), and muscle artifact. Subsequently, for each segment, the bad channels have their data interpolated with spherical splines, as in FASTER. This allows users to maintain the maximum number of available segments, while still maximizing artifact rejection within individual segments.
  10. captures both frequency and location information Artifacts should be decomposable in a wavelet basis. Mother wavelet, the shrinkage rule, and the threshold are important to the design of the noise removal method (Daly et al. 2013).
  11. This problem is widely studied and large number of
  12. Annotation or reference channel
  13. Annotation or reference channel
  14. In this paper, we present a novel software-hardware system that uses a weak supervisory signal to indicate that some noise is occurring, but not what the source of the noise is or how it is manifested in the EEG signal. The slide shows our proposed systtem architecture , First First 64 EEG data are decomposed into independent components using ICA, and these components form bags that are labeled 1/0 where 1 indicates that there is a noise. Each bag is then classified by a multi-instance learning algorithm to identify if it has noise component or not. We use Symoblic Aggregate Approximation (SAX) for featurizing each component. We go through each of these subsystems in next slides
  15. https://www.slideshare.net/yokotatsuya/independent-component-analysis-11359849/3
  16. This problem is widely studied and large number of
  17. https://www.slideshare.net/yokotatsuya/independent-component-analysis-11359849/3
  18. https://www.slideshare.net/yokotatsuya/independent-component-analysis-11359849/3
  19. In this paper, we present a novel software-hardware system that uses a weak supervisory signal to indicate that some noise is occurring, but not what the source of the noise is or how it is manifested in the EEG signal. The slide shows our proposed systtem architecture , First First 64 EEG data are decomposed into independent components using ICA, and these components form bags that are labeled 1/0 where 1 indicates that there is a noise. Each bag is then classified by a multi-instance learning algorithm to identify if it has noise component or not. We use Symoblic Aggregate Approximation (SAX) for featurizing each component. We go through each of these subsystems in next slides
  20. We featurize each component using SAX before using multi instance learning. This slide demonstrates how the SAX works on a simple time series signal. We partition the signal into different levels where each level manifests a letter. For example, this window of the signal is partitioned into three levels which represents the symbols of “baabccbc” word. We use bag of words generated from SAX to classify the time series. For featurizing EEG, we convert time series to symbols In this case time series shown in first figure is converted to symbols “baabccbc”.
  21. In this paper, we present a novel software-hardware system that uses a weak supervisory signal to indicate that some noise is occurring, but not what the source of the noise is or how it is manifested in the EEG signal. The slide shows our proposed systtem architecture , First First 64 EEG data are decomposed into independent components using ICA, and these components form bags that are labeled 1/0 where 1 indicates that there is a noise. Each bag is then classified by a multi-instance learning algorithm to identify if it has noise component or not. We use Symoblic Aggregate Approximation (SAX) for featurizing each component. We go through each of these subsystems in next slides
  22. Our goal is given a new bag predict if it is positive or negative.
  23. Multi instance learning algorithms also allows us to predict if the instance is positive or negative. In our case positive instance corresponds to noisy component in EEG data.
  24. In this paper, we present a novel software-hardware system that uses a weak supervisory signal to indicate that some noise is occurring, but not what the source of the noise is or how it is manifested in the EEG signal. The slide shows our proposed systtem architecture , First First 64 EEG data are decomposed into independent components using ICA, and these components form bags that are labeled 1/0 where 1 indicates that there is a noise. Each bag is then classified by a multi-instance learning algorithm to identify if it has noise component or not. We use Symoblic Aggregate Approximation (SAX) for featurizing each component. We go through each of these subsystems in next slides
  25. Although few reasons these numbers cannot be directly compared: We do 3-Fold cross validation, they randomly generate 20 train test split They are using 7 patients from same dataset. I am not sure which patients they are using The goals are different they are doing classification, we are doing classification as well as artifact component removal. Also, I know there are problems with evaluation, I will talk about evaluation at the end
  26. Another approach we are working on to remove artifacts from noisy EEG data is using generative adversarial networks. This approach has been used recently in computer vision for converting images from one domain to another. For example, in this paper, unpaired image-to-image translation using cycle-consistent adversarial networks (CycleGAN) convert paintings of Monet to photos. They also convert photos of zebras to horses, winter to summer and reverse. We want to use this deep neural network to remove artifacts from EEG data. This architecture is a natural fit for this problem as it does not require any paired data. I will go over how cycleGAN work in the next few slides.
  27. For explaining cycleGAN, I will use example of converting a horse image to zebra image as it is easier to explain. In this example, X is images of all horses and Y is images of all zebras. G is convolutional autoencoder which converts horse image to corresponding zebra image and F is another convolutional autoencoder which converts zebra image to horse image. Dx and Dy are convolutional neural networks that discriminate between generated and real images. For training this neural network, we give a horse and zebra image as input and calculate the loss. After calculating the loss, we use standard gradient descent for backpropagation and train the network. We describe how we calculate the loss function in next slide.
  28. Another approach we are working on to remove artifacts from noisy EEG data is using generative adversarial networks. This approach has been used recently in computer vision for converting images from one domain to another. For example, in this paper, unpaired image-to-image translation using cycle-consistent adversarial networks (CycleGAN) convert paintings of Monet to photos. They also convert photos of zebras to horses, winter to summer and reverse. We want to use this deep neural network to remove artifacts from EEG data. This architecture is a natural fit for this problem as it does not require any paired data. I will go over how cycleGAN work in the next few slides.
  29. Give Example Talk about how network cheats
  30. Visual Analogy of the problem
  31. Give Example Talk about how network cheats
  32. Give Example Talk about how network cheats
  33. Color the arrows
  34. To train the asymmetricGAN and reduce model oscillation, we use the strategy of (Shrivastava et al. 2016) of saving the history of generated time series. We update the discriminators using a history of generated signals rather than the ones produced by the latest generative networks. We keep a buffer that stores the 50 previously generated signals for training. In the next section, we show the effectiveness of our method on synthetic and EEG datasets. Explain the architecture visually
  35. Artifact removal from EEG data is a problem where ground truth does not exist be- cause we do not have a clean version of the noisy signal. This makes the evaluation of artifact removal methods difficult. Also, visualizing and understanding EEG data is a time consuming task. We solve this problem by creating a simpler synthetic dataset.
  36. Talk about the fact that this is qualitative example how mse just gives consistency across the dataset MSE(a,b) = 0.5
  37. Talk about the fact that this is qualitative example how mse just gives consistency across the dataset MSE(a,b) = 0.5
  38. Give Example
  39. More Examples in Thesis
  40. Qualitative Evaluation Subjective Useful for validating the algorithm.
  41. Most used Evaluation using detection
  42. Most used Methods generally differ in how they model characteristics of artifact signals. (Daly et al. 2012) proposes to use mean power and its standard deviation, the maximum amplitude and its standard deviation, kurtosis, and skewness of the amplitude of resting state background rhythms to characterize the clean and noisy EEG signal. (Mognon et al. 2010) learns features such as kur- tosis, spatial average difference, and maximum epoch variance for modeling ocular artifacts. (Delorme, Sejnowski, and Makeig 2007b) learns to characterize a larger variety of artifacts using extreme values, linear trends, data improbability, kurtosis and spectral patterns. We use an artifact detector that uses a convolutional neural network to evaluate artifact removal algorithms in Chapter 4.
  43. Most used Methods generally differ in how they model characteristics of artifact signals. (Daly et al. 2012) proposes to use mean power and its standard deviation, the maximum amplitude and its standard deviation, kurtosis, and skewness of the amplitude of resting state background rhythms to characterize the clean and noisy EEG signal. (Mognon et al. 2010) learns features such as kur- tosis, spatial average difference, and maximum epoch variance for modeling ocular artifacts. (Delorme, Sejnowski, and Makeig 2007b) learns to characterize a larger variety of artifacts using extreme values, linear trends, data improbability, kurtosis and spectral patterns. We use an artifact detector that uses a convolutional neural network to evaluate artifact removal algorithms in Chapter 4.
  44. Most used Multiple evaluation methods for artifact removal algorithms have been proposed in the literature. Each of these methods has their strengths and limitations. Thus, it is important to use as many evaluation mechanisms as possible. But, most existing papers use one or two of these methods to evaluate their algorithm. This makes performance comparison of these methods hard. This problem is exacerbated because several variants of each of these methods exist in literature and they are tested on different datasets. Our goal in this thesis is to standardize the evaluation mechanism and create a tool for evaluating artifact removal algorithms using all of the above approaches. We also intend to use this tool for comparing our approach with existing methods discussed above.
  45. where speech can be denoised before amplification.
  46. TODO : Reduce content on slide 65