How can we classify the EEG signals By using Machine Learning Algorithms ?
Hi
I have explained the stages:
First, for the raw EEG signals, first you need to apply pre-processing tasks like feature extraction. Various methods are available for the same like VLC methods, PCA or wavelets etc
Then apply classification methods over the transformed data.
Generally if you apply classification algorithms directly on the raw data, then it does not yield good accuracy and good models.
I have added some links to research articles free PDFs on the same:
E. Parvinnia, M. Sabeti, M. Zolghadri Jahromi, R. Boostani, Classification of EEG Signals using adaptive weighted distance nearest neighbor algorithm, Journal of King Saud University - Computer and Information Sciences, Volume 26, Issue 1, January 2014, Pages 1-6, ISSN 1319-1578, http://dx.doi.org/10.1016/j.jksuci.2013.01.001.
(http://www.sciencedirect.com/science/article/pii/S1319157813000025)
Keywords: Nearest neighbor; Noisy training data; EEG signal classification; Band power; Fractal dimension; Autoregressive coefficients
Nazre Abdul Rashid, Mohd. Nasir Taib, Sahrim Lias, Norizam Sulaiman, Zunairah Hj. Murat, Ros Shilawani S. Abdul Kadir, Learners’ Learning Style Classification related to IQ and Stress based on EEG, Procedia - Social and Behavioral Sciences, Volume 29, 2011, Pages 1061-1070, ISSN 1877-0428, http://dx.doi.org/10.1016/j.sbspro.2011.11.339.
(http://www.sciencedirect.com/science/article/pii/S187704281102800X)
Keywords: Learning Style; EEG; IQ; Stress; Classification
Farnaz Ghassemi, Mohammad Hassan_Moradi, Mehdi Tehrani-Doost, Vahid Abootalebi, Using non-linear features of EEG for ADHD/normal participants’ classification, Procedia - Social and Behavioral Sciences, Volume 32, 2012, Pages 148-152, ISSN 1877-0428, http://dx.doi.org/10.1016/j.sbspro.2012.01.024.
(http://www.sciencedirect.com/science/article/pii/S1877042812000250)
Keywords: Attention-Deficit Hyperactivity Disorder (ADHD); K-nearest neighbors classifier; Continuous Performance Test (CPT); electroencephalogram; feature extraction; sustained attention
Vernon Lawhern, W. David Hairston, Kaleb McDowell, Marissa Westerfield, Kay Robbins, Detection and classification of subject-generated artifacts in EEG signals using autoregressive models, Journal of Neuroscience Methods, Volume 208, Issue 2, 15 July 2012, Pages 181-189, ISSN 0165-0270, http://dx.doi.org/10.1016/j.jneumeth.2012.05.017.
(http://www.sciencedirect.com/science/article/pii/S0165027012001860)
Keywords: Autoregressive model; Artifacts; Electroencephalography; Support vector machines
T. Radüntz, J. Scouten, O. Hochmuth, B. Meffert, EEG artifact elimination by extraction of ICA-component features using image processing algorithms, Journal of Neuroscience Methods, Volume 243, 30 March 2015, Pages 84-93, ISSN 0165-0270, http://dx.doi.org/10.1016/j.jneumeth.2015.01.030.
(http://www.sciencedirect.com/science/article/pii/S0165027015000370)
Keywords: Independent component analysis; ICA; EEG; Artifact elimination; Image processing; Local binary patterns; Range filter; Geometric features
P. Bhuvaneswari, J. Satheesh Kumar, Influence of Linear Features in Nonlinear Electroencephalography (EEG) Signals, Procedia Computer Science, Volume 47, 2015, Pages 229-236, ISSN 1877-0509, http://dx.doi.org/10.1016/j.procs.2015.03.202.
(http://www.sciencedirect.com/science/article/pii/S1877050915004706)
Keywords: Classifier; EEG; Majority rule; RMS; Variance; LPC ;
Ericka Janet Rechy-Ramirez, Huosheng Hu, Bio-signal based control in assistive robots: a survey, Digital Communications and Networks, Volume 1, Issue 2, April 2015, Pages 85-101, ISSN 2352-8648, http://dx.doi.org/10.1016/j.dcan.2015.02.004.
(http://www.sciencedirect.com/science/article/pii/S2352864815000097)
Keywords: Assistive robots; EMG; EEG; Feature extraction and classification
Zarita Zainuddin, Lai Kee Huong, Ong Pauline, On the Use of Wavelet Neural Networks in the Task of Epileptic Seizure Detection from Electroencephalography Signals, Procedia Computer Science, Volume 11, 2012, Pages 149-159, ISSN 1877-0509, http://dx.doi.org/10.1016/j.procs.2012.09.016.
(http://www.sciencedirect.com/science/article/pii/S1877050912005996)
Keywords: EEG signals; epilepsy; epileptic seizure; wavelet neural networks; wavelet transform
Nikita Gurudath, H. Bryan Riley, Drowsy Driving Detection by EEG Analysis Using Wavelet Transform and K-means Clustering, Procedia Computer Science, Volume 34, 2014, Pages 400-409, ISSN 1877-0509, http://dx.doi.org/10.1016/j.procs.2014.07.045.
(http://www.sciencedirect.com/science/article/pii/S1877050914009004)
Keywords: Driver health monitoring; EEG signal analysis; discrete wavelet transform; Debauchies’ wavelet; K-means clustering
Vladimir V. Galatenko, Eugene D. Livshitz, Vladimir M. Staroverov, Taras P. Lukashenko, Alexey A. Galatenko, Vladimir E. Podol'skii, Victor A. Sadovnichy, Vyacheslav V. Lebedev, Sergey A. Isaychev, Alexandr M. Chernorizov, Yuriy P. Zinchenko, A Remark on the Most Informative EEG Signal Components in a Super-scalable Method for Functional State Classification based on the Wavelet Decomposition, Procedia - Social and Behavioral Sciences, Volume 86, 10 October 2013, Pages 18-23, ISSN 1877-0428, http://dx.doi.org/10.1016/j.sbspro.2013.08.518.
(http://www.sciencedirect.com/science/article/pii/S1877042813026499)
Keywords: Functional state; Stress; Calm wakefulness; Automated classification; electroencephalogram; Wavelet decomposition; CDF wavelets; Frequency range; Frequency localization.
R.M. Isa, I. Pasya, M.N. Taib, A.H. Jahidin, W.R.W. Omar, N. Fuad, H. Norhazman, S.B. Kutty, S.F.S. Adnan, Classification of Brainwave Asymmetry Influenced by Mobile Phone Radiofrequency Emission, Procedia - Social and Behavioral Sciences, Volume 97, 6 November 2013, Pages 538-545, ISSN 1877-0428, http://dx.doi.org/10.1016/j.sbspro.2013.10.270.
(http://www.sciencedirect.com/science/article/pii/S1877042813037154)
Keywords: EEG; alpha sub-band; asymmetry; RF; mobile phone radiation
Wafaa Khazaal Shams, Abdul Wahab, Imad Fakhri, Affective Computing Model Using Source-temporal Domain, Procedia - Social and Behavioral Sciences, Volume 97, 6 November 2013, Pages 54-62, ISSN 1877-0428, http://dx.doi.org/10.1016/j.sbspro.2013.10.204.
(http://www.sciencedirect.com/science/article/pii/S1877042813036495)
Keywords: 2 Dimension affective model; EEG; emotion; relative source temporal features; RafD faces
Marini Othman, Abdul Wahab, Izzah Karim, Mariam Adawiah Dzulkifli, Imad Fakhri Taha Alshaikli, EEG Emotion Recognition Based on the Dimensional Models of Emotions, Procedia - Social and Behavioral Sciences, Volume 97, 6 November 2013, Pages 30-37, ISSN 1877-0428, http://dx.doi.org/10.1016/j.sbspro.2013.10.201.
(http://www.sciencedirect.com/science/article/pii/S187704281303646X)
Keywords: brain signals; valence-arousal model; preschoolers; children; emotions; machine learning; classification.
Prashant Lahane, Arun Kumar Sangaiah, An Approach to EEG Based Emotion Recognition and Classification Using Kernel Density Estimation, Procedia Computer Science, Volume 48, 2015, Pages 574-581, ISSN 1877-0509, http://dx.doi.org/10.1016/j.procs.2015.04.138.
(http://www.sciencedirect.com/science/article/pii/S187705091500647X)
Keywords: Electroencephalography (EEG) Signals; Independent Component Analysis (ICA); Kernel Density Estimation (KDE); Artificial Neural Network (ANN)
Satchidanada Dehuri, Alok Kumar Jagadev, Sung-Bae Cho, Epileptic Seizure Identification from Electroencephalography Signal Using DE-RBFNs Ensemble, Procedia Computer Science, Volume 23, 2013, Pages 84-95, ISSN 1877-0509, http://dx.doi.org/10.1016/j.procs.2013.10.012.
(http://www.sciencedirect.com/science/article/pii/S1877050913011472)
Keywords: EEG; classification; radial basis function neural networks; differential evolution; bagging
Z. Elahi, R. Boostani, A. Motie Nasrabadi, Estimation of hypnosis susceptibility based on electroencephalogram signal features, Scientia Iranica, Volume 20, Issue 3, June 2013, Pages 730-737, ISSN 1026-3098, http://dx.doi.org/10.1016/j.scient.2012.07.015.
(http://www.sciencedirect.com/science/article/pii/S1026309812001927)
Keywords: Degrees of hypnotic susceptibility; EEG; Classification; Fuzzy nearest neighbor; Fuzzy Rule-based Classification System (FRBCS)
Joseph C. McBride, Xiaopeng Zhao, Nancy B. Munro, Gregory A. Jicha, Frederick A. Schmitt, Richard J. Kryscio, Charles D. Smith, Yang Jiang, Sugihara causality analysis of scalp EEG for detection of early Alzheimer's disease, NeuroImage: Clinical, Volume 7, 2015, Pages 258-265, ISSN 2213-1582, http://dx.doi.org/10.1016/j.nicl.2014.12.005.
(http://www.sciencedirect.com/science/article/pii/S2213158214001909)
Keywords: Early Alzheimer's disease; Mild cognitive impairment; EEG-based diagnosis; Causality analysis
Hi,
It depends what you want to accomplish. I would use neural networks for example:
tensorflow.org or https://github.com/mila-udem/blocks
Easier solution would be using Mathlab with DeepLearnToolbox
But also:
-Generalised Linear Models
-Hierarchical Bayesian Models: In order to link up the idea of evidence accumulation to EEG power spectrum data
-Information theoretic measures applied with AutoRegression models to understand consciousness
-SVMs used when the task is a 2-Alternative Forced Choice as a classification measure.
thnanks for reply Krzysztof wolk. actually i want to classify the signals as normal and abnormal for seizure prediction
I think that you can use a signal stream or EPOCH and define the characteristic that defines the feature to be different from each other (e.g. their value over a certain period of time). Maybe if there is a different energy (or total power value) you can use this feature and feed these vectors ( I presume you have different training samples) to your training (in my case Supervised Learning Methods worked out fine with EEG processed signals). If you one you can read one paper of mine to see how I performed some classification, maybe it is useful for you.
Thanks for your time on reading my reply
Hi
I have explained the stages:
First, for the raw EEG signals, first you need to apply pre-processing tasks like feature extraction. Various methods are available for the same like VLC methods, PCA or wavelets etc
Then apply classification methods over the transformed data.
Generally if you apply classification algorithms directly on the raw data, then it does not yield good accuracy and good models.
I have added some links to research articles free PDFs on the same:
E. Parvinnia, M. Sabeti, M. Zolghadri Jahromi, R. Boostani, Classification of EEG Signals using adaptive weighted distance nearest neighbor algorithm, Journal of King Saud University - Computer and Information Sciences, Volume 26, Issue 1, January 2014, Pages 1-6, ISSN 1319-1578, http://dx.doi.org/10.1016/j.jksuci.2013.01.001.
(http://www.sciencedirect.com/science/article/pii/S1319157813000025)
Keywords: Nearest neighbor; Noisy training data; EEG signal classification; Band power; Fractal dimension; Autoregressive coefficients
Nazre Abdul Rashid, Mohd. Nasir Taib, Sahrim Lias, Norizam Sulaiman, Zunairah Hj. Murat, Ros Shilawani S. Abdul Kadir, Learners’ Learning Style Classification related to IQ and Stress based on EEG, Procedia - Social and Behavioral Sciences, Volume 29, 2011, Pages 1061-1070, ISSN 1877-0428, http://dx.doi.org/10.1016/j.sbspro.2011.11.339.
(http://www.sciencedirect.com/science/article/pii/S187704281102800X)
Keywords: Learning Style; EEG; IQ; Stress; Classification
Farnaz Ghassemi, Mohammad Hassan_Moradi, Mehdi Tehrani-Doost, Vahid Abootalebi, Using non-linear features of EEG for ADHD/normal participants’ classification, Procedia - Social and Behavioral Sciences, Volume 32, 2012, Pages 148-152, ISSN 1877-0428, http://dx.doi.org/10.1016/j.sbspro.2012.01.024.
(http://www.sciencedirect.com/science/article/pii/S1877042812000250)
Keywords: Attention-Deficit Hyperactivity Disorder (ADHD); K-nearest neighbors classifier; Continuous Performance Test (CPT); electroencephalogram; feature extraction; sustained attention
Vernon Lawhern, W. David Hairston, Kaleb McDowell, Marissa Westerfield, Kay Robbins, Detection and classification of subject-generated artifacts in EEG signals using autoregressive models, Journal of Neuroscience Methods, Volume 208, Issue 2, 15 July 2012, Pages 181-189, ISSN 0165-0270, http://dx.doi.org/10.1016/j.jneumeth.2012.05.017.
(http://www.sciencedirect.com/science/article/pii/S0165027012001860)
Keywords: Autoregressive model; Artifacts; Electroencephalography; Support vector machines
T. Radüntz, J. Scouten, O. Hochmuth, B. Meffert, EEG artifact elimination by extraction of ICA-component features using image processing algorithms, Journal of Neuroscience Methods, Volume 243, 30 March 2015, Pages 84-93, ISSN 0165-0270, http://dx.doi.org/10.1016/j.jneumeth.2015.01.030.
(http://www.sciencedirect.com/science/article/pii/S0165027015000370)
Keywords: Independent component analysis; ICA; EEG; Artifact elimination; Image processing; Local binary patterns; Range filter; Geometric features
P. Bhuvaneswari, J. Satheesh Kumar, Influence of Linear Features in Nonlinear Electroencephalography (EEG) Signals, Procedia Computer Science, Volume 47, 2015, Pages 229-236, ISSN 1877-0509, http://dx.doi.org/10.1016/j.procs.2015.03.202.
(http://www.sciencedirect.com/science/article/pii/S1877050915004706)
Keywords: Classifier; EEG; Majority rule; RMS; Variance; LPC ;
Ericka Janet Rechy-Ramirez, Huosheng Hu, Bio-signal based control in assistive robots: a survey, Digital Communications and Networks, Volume 1, Issue 2, April 2015, Pages 85-101, ISSN 2352-8648, http://dx.doi.org/10.1016/j.dcan.2015.02.004.
(http://www.sciencedirect.com/science/article/pii/S2352864815000097)
Keywords: Assistive robots; EMG; EEG; Feature extraction and classification
Zarita Zainuddin, Lai Kee Huong, Ong Pauline, On the Use of Wavelet Neural Networks in the Task of Epileptic Seizure Detection from Electroencephalography Signals, Procedia Computer Science, Volume 11, 2012, Pages 149-159, ISSN 1877-0509, http://dx.doi.org/10.1016/j.procs.2012.09.016.
(http://www.sciencedirect.com/science/article/pii/S1877050912005996)
Keywords: EEG signals; epilepsy; epileptic seizure; wavelet neural networks; wavelet transform
Nikita Gurudath, H. Bryan Riley, Drowsy Driving Detection by EEG Analysis Using Wavelet Transform and K-means Clustering, Procedia Computer Science, Volume 34, 2014, Pages 400-409, ISSN 1877-0509, http://dx.doi.org/10.1016/j.procs.2014.07.045.
(http://www.sciencedirect.com/science/article/pii/S1877050914009004)
Keywords: Driver health monitoring; EEG signal analysis; discrete wavelet transform; Debauchies’ wavelet; K-means clustering
Vladimir V. Galatenko, Eugene D. Livshitz, Vladimir M. Staroverov, Taras P. Lukashenko, Alexey A. Galatenko, Vladimir E. Podol'skii, Victor A. Sadovnichy, Vyacheslav V. Lebedev, Sergey A. Isaychev, Alexandr M. Chernorizov, Yuriy P. Zinchenko, A Remark on the Most Informative EEG Signal Components in a Super-scalable Method for Functional State Classification based on the Wavelet Decomposition, Procedia - Social and Behavioral Sciences, Volume 86, 10 October 2013, Pages 18-23, ISSN 1877-0428, http://dx.doi.org/10.1016/j.sbspro.2013.08.518.
(http://www.sciencedirect.com/science/article/pii/S1877042813026499)
Keywords: Functional state; Stress; Calm wakefulness; Automated classification; electroencephalogram; Wavelet decomposition; CDF wavelets; Frequency range; Frequency localization.
R.M. Isa, I. Pasya, M.N. Taib, A.H. Jahidin, W.R.W. Omar, N. Fuad, H. Norhazman, S.B. Kutty, S.F.S. Adnan, Classification of Brainwave Asymmetry Influenced by Mobile Phone Radiofrequency Emission, Procedia - Social and Behavioral Sciences, Volume 97, 6 November 2013, Pages 538-545, ISSN 1877-0428, http://dx.doi.org/10.1016/j.sbspro.2013.10.270.
(http://www.sciencedirect.com/science/article/pii/S1877042813037154)
Keywords: EEG; alpha sub-band; asymmetry; RF; mobile phone radiation
Wafaa Khazaal Shams, Abdul Wahab, Imad Fakhri, Affective Computing Model Using Source-temporal Domain, Procedia - Social and Behavioral Sciences, Volume 97, 6 November 2013, Pages 54-62, ISSN 1877-0428, http://dx.doi.org/10.1016/j.sbspro.2013.10.204.
(http://www.sciencedirect.com/science/article/pii/S1877042813036495)
Keywords: 2 Dimension affective model; EEG; emotion; relative source temporal features; RafD faces
Marini Othman, Abdul Wahab, Izzah Karim, Mariam Adawiah Dzulkifli, Imad Fakhri Taha Alshaikli, EEG Emotion Recognition Based on the Dimensional Models of Emotions, Procedia - Social and Behavioral Sciences, Volume 97, 6 November 2013, Pages 30-37, ISSN 1877-0428, http://dx.doi.org/10.1016/j.sbspro.2013.10.201.
(http://www.sciencedirect.com/science/article/pii/S187704281303646X)
Keywords: brain signals; valence-arousal model; preschoolers; children; emotions; machine learning; classification.
Prashant Lahane, Arun Kumar Sangaiah, An Approach to EEG Based Emotion Recognition and Classification Using Kernel Density Estimation, Procedia Computer Science, Volume 48, 2015, Pages 574-581, ISSN 1877-0509, http://dx.doi.org/10.1016/j.procs.2015.04.138.
(http://www.sciencedirect.com/science/article/pii/S187705091500647X)
Keywords: Electroencephalography (EEG) Signals; Independent Component Analysis (ICA); Kernel Density Estimation (KDE); Artificial Neural Network (ANN)
Satchidanada Dehuri, Alok Kumar Jagadev, Sung-Bae Cho, Epileptic Seizure Identification from Electroencephalography Signal Using DE-RBFNs Ensemble, Procedia Computer Science, Volume 23, 2013, Pages 84-95, ISSN 1877-0509, http://dx.doi.org/10.1016/j.procs.2013.10.012.
(http://www.sciencedirect.com/science/article/pii/S1877050913011472)
Keywords: EEG; classification; radial basis function neural networks; differential evolution; bagging
Z. Elahi, R. Boostani, A. Motie Nasrabadi, Estimation of hypnosis susceptibility based on electroencephalogram signal features, Scientia Iranica, Volume 20, Issue 3, June 2013, Pages 730-737, ISSN 1026-3098, http://dx.doi.org/10.1016/j.scient.2012.07.015.
(http://www.sciencedirect.com/science/article/pii/S1026309812001927)
Keywords: Degrees of hypnotic susceptibility; EEG; Classification; Fuzzy nearest neighbor; Fuzzy Rule-based Classification System (FRBCS)
Joseph C. McBride, Xiaopeng Zhao, Nancy B. Munro, Gregory A. Jicha, Frederick A. Schmitt, Richard J. Kryscio, Charles D. Smith, Yang Jiang, Sugihara causality analysis of scalp EEG for detection of early Alzheimer's disease, NeuroImage: Clinical, Volume 7, 2015, Pages 258-265, ISSN 2213-1582, http://dx.doi.org/10.1016/j.nicl.2014.12.005.
(http://www.sciencedirect.com/science/article/pii/S2213158214001909)
Keywords: Early Alzheimer's disease; Mild cognitive impairment; EEG-based diagnosis; Causality analysis
This is indeed very informative Dr. Mandal, though I did not post the question originally. If I want to do some study with my students can you recommend some datasets of EEG as well. Thank you in advance.
see this:https://www.researchgate.net/publication/301325717_Improved_EEG_Event_Classification_Using_Differential_Energy
Conference Paper Improved EEG Event Classification Using Differential Energy
- Badges
- Science topic