Using Morphological-Linear Neural Network for Upper Limb Movement Intention Recognition from EEG Signals

Gerardo Hernández; Luis G. Hernández; Erik Zamora; Humberto Sossa; Javier M. Antelis; Omar Mendoza-Montoya; Luis E. Falcón

doi:10.1007/978-3-030-21077-9_36

Using Morphological-Linear Neural Network for Upper Limb Movement Intention Recognition from EEG Signals

Gerardo Hernández, Luis G. Hernández, Erik Zamora, Humberto Sossa, Javier M. Antelis, Omar Mendoza-Montoya, Luis E. Falcón

Centro de Investigación en Computación (CIC)

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

1 Scopus citations

Abstract

This study aims to compare classical and Morphological-Linear Neural Network (MLNN) algorithms for the intention recognition to perform different movements from electroencephalographic (EEG) signals. Three classification models were implemented and assessed to decode EEG motor imagery signals: (i) Morphological-Linear Neural Network (MLNN) (ii) Support Vector Machine (SVM) and (iii) Multilayer perceptron (MLP). Real EEG signals recorded during robot-assisted rehabilitation therapy were used to evaluate the performance of the proposed algorithms in the classification of three classes (relax, movement intention A Int A and movement intention B Int B) using multi-CSP based features extracted from EEG signals. The results of a ten-fold cross validation show similar results in terms of classification accuracy for the SVM and MLNN models. However, the number of parameters used in each model varies considerably (the MLNN model use less parameters than the SVM). This study indicates potential application of MLNNs for decoding movement intentions and its use to develop more natural and intuitive robot assisted neurorehabilitation therapies.

Original language	English
Title of host publication	Pattern Recognition - 11th Mexican Conference, MCPR 2019, Proceedings
Editors	Jesús Ariel Carrasco-Ochoa, José Francisco Martínez-Trinidad, José Arturo Olvera-López, Joaquín Salas
Publisher	Springer Verlag
Pages	389-397
Number of pages	9
ISBN (Print)	9783030210762
DOIs	https://doi.org/10.1007/978-3-030-21077-9_36
State	Published - 2019
Event	11th Mexican Conference on Pattern Recognition, MCPR 2019 - Querétaro, Mexico Duration: 26 Jun 2019 → 29 Jun 2019

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	11524 LNCS
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	11th Mexican Conference on Pattern Recognition, MCPR 2019
Country/Territory	Mexico
City	Querétaro
Period	26/06/19 → 29/06/19

Keywords

Brain-computer interfaces
Electroencephalogram
Machine learning
Morphological-linear neural network
Movement planing

Access to Document

10.1007/978-3-030-21077-9_36

Cite this

Hernández, G., Hernández, L. G., Zamora, E., Sossa, H., Antelis, J. M., Mendoza-Montoya, O., & Falcón, L. E. (2019). Using Morphological-Linear Neural Network for Upper Limb Movement Intention Recognition from EEG Signals. In J. A. Carrasco-Ochoa, J. F. Martínez-Trinidad, J. A. Olvera-López, & J. Salas (Eds.), Pattern Recognition - 11th Mexican Conference, MCPR 2019, Proceedings (pp. 389-397). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 11524 LNCS). Springer Verlag. https://doi.org/10.1007/978-3-030-21077-9_36

Hernández, Gerardo ; Hernández, Luis G. ; Zamora, Erik et al. / Using Morphological-Linear Neural Network for Upper Limb Movement Intention Recognition from EEG Signals. Pattern Recognition - 11th Mexican Conference, MCPR 2019, Proceedings. editor / Jesús Ariel Carrasco-Ochoa ; José Francisco Martínez-Trinidad ; José Arturo Olvera-López ; Joaquín Salas. Springer Verlag, 2019. pp. 389-397 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{030f8e115c2c49768cde3e1c3c206e90,

title = "Using Morphological-Linear Neural Network for Upper Limb Movement Intention Recognition from EEG Signals",

abstract = "This study aims to compare classical and Morphological-Linear Neural Network (MLNN) algorithms for the intention recognition to perform different movements from electroencephalographic (EEG) signals. Three classification models were implemented and assessed to decode EEG motor imagery signals: (i) Morphological-Linear Neural Network (MLNN) (ii) Support Vector Machine (SVM) and (iii) Multilayer perceptron (MLP). Real EEG signals recorded during robot-assisted rehabilitation therapy were used to evaluate the performance of the proposed algorithms in the classification of three classes (relax, movement intention A Int A and movement intention B Int B) using multi-CSP based features extracted from EEG signals. The results of a ten-fold cross validation show similar results in terms of classification accuracy for the SVM and MLNN models. However, the number of parameters used in each model varies considerably (the MLNN model use less parameters than the SVM). This study indicates potential application of MLNNs for decoding movement intentions and its use to develop more natural and intuitive robot assisted neurorehabilitation therapies.",

keywords = "Brain-computer interfaces, Electroencephalogram, Machine learning, Morphological-linear neural network, Movement planing",

author = "Gerardo Hern{\'a}ndez and Hern{\'a}ndez, {Luis G.} and Erik Zamora and Humberto Sossa and Antelis, {Javier M.} and Omar Mendoza-Montoya and Falc{\'o}n, {Luis E.}",

note = "Publisher Copyright: {\textcopyright} 2019, Springer Nature Switzerland AG.; 11th Mexican Conference on Pattern Recognition, MCPR 2019 ; Conference date: 26-06-2019 Through 29-06-2019",

year = "2019",

doi = "10.1007/978-3-030-21077-9_36",

language = "Ingl{\'e}s",

isbn = "9783030210762",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

publisher = "Springer Verlag",

pages = "389--397",

editor = "Carrasco-Ochoa, {Jes{\'u}s Ariel} and Mart{\'i}nez-Trinidad, {Jos{\'e} Francisco} and Olvera-L{\'o}pez, {Jos{\'e} Arturo} and Joaqu{\'i}n Salas",

booktitle = "Pattern Recognition - 11th Mexican Conference, MCPR 2019, Proceedings",

address = "Alemania",

}

Hernández, G, Hernández, LG, Zamora, E , Sossa, H, Antelis, JM, Mendoza-Montoya, O & Falcón, LE 2019, Using Morphological-Linear Neural Network for Upper Limb Movement Intention Recognition from EEG Signals. in JA Carrasco-Ochoa, JF Martínez-Trinidad, JA Olvera-López & J Salas (eds), Pattern Recognition - 11th Mexican Conference, MCPR 2019, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 11524 LNCS, Springer Verlag, pp. 389-397, 11th Mexican Conference on Pattern Recognition, MCPR 2019, Querétaro, Mexico, 26/06/19. https://doi.org/10.1007/978-3-030-21077-9_36

Using Morphological-Linear Neural Network for Upper Limb Movement Intention Recognition from EEG Signals. / Hernández, Gerardo; Hernández, Luis G.; Zamora, Erik et al.
Pattern Recognition - 11th Mexican Conference, MCPR 2019, Proceedings. ed. / Jesús Ariel Carrasco-Ochoa; José Francisco Martínez-Trinidad; José Arturo Olvera-López; Joaquín Salas. Springer Verlag, 2019. p. 389-397 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 11524 LNCS).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Using Morphological-Linear Neural Network for Upper Limb Movement Intention Recognition from EEG Signals

AU - Hernández, Gerardo

AU - Hernández, Luis G.

AU - Zamora, Erik

AU - Sossa, Humberto

AU - Antelis, Javier M.

AU - Mendoza-Montoya, Omar

AU - Falcón, Luis E.

PY - 2019

Y1 - 2019

N2 - This study aims to compare classical and Morphological-Linear Neural Network (MLNN) algorithms for the intention recognition to perform different movements from electroencephalographic (EEG) signals. Three classification models were implemented and assessed to decode EEG motor imagery signals: (i) Morphological-Linear Neural Network (MLNN) (ii) Support Vector Machine (SVM) and (iii) Multilayer perceptron (MLP). Real EEG signals recorded during robot-assisted rehabilitation therapy were used to evaluate the performance of the proposed algorithms in the classification of three classes (relax, movement intention A Int A and movement intention B Int B) using multi-CSP based features extracted from EEG signals. The results of a ten-fold cross validation show similar results in terms of classification accuracy for the SVM and MLNN models. However, the number of parameters used in each model varies considerably (the MLNN model use less parameters than the SVM). This study indicates potential application of MLNNs for decoding movement intentions and its use to develop more natural and intuitive robot assisted neurorehabilitation therapies.

AB - This study aims to compare classical and Morphological-Linear Neural Network (MLNN) algorithms for the intention recognition to perform different movements from electroencephalographic (EEG) signals. Three classification models were implemented and assessed to decode EEG motor imagery signals: (i) Morphological-Linear Neural Network (MLNN) (ii) Support Vector Machine (SVM) and (iii) Multilayer perceptron (MLP). Real EEG signals recorded during robot-assisted rehabilitation therapy were used to evaluate the performance of the proposed algorithms in the classification of three classes (relax, movement intention A Int A and movement intention B Int B) using multi-CSP based features extracted from EEG signals. The results of a ten-fold cross validation show similar results in terms of classification accuracy for the SVM and MLNN models. However, the number of parameters used in each model varies considerably (the MLNN model use less parameters than the SVM). This study indicates potential application of MLNNs for decoding movement intentions and its use to develop more natural and intuitive robot assisted neurorehabilitation therapies.

KW - Brain-computer interfaces

KW - Electroencephalogram

KW - Machine learning

KW - Morphological-linear neural network

KW - Movement planing

UR - http://www.scopus.com/inward/record.url?scp=85068335278&partnerID=8YFLogxK

U2 - 10.1007/978-3-030-21077-9_36

DO - 10.1007/978-3-030-21077-9_36

M3 - Contribución a la conferencia

AN - SCOPUS:85068335278

SN - 9783030210762

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 389

EP - 397

BT - Pattern Recognition - 11th Mexican Conference, MCPR 2019, Proceedings

A2 - Carrasco-Ochoa, Jesús Ariel

A2 - Martínez-Trinidad, José Francisco

A2 - Olvera-López, José Arturo

A2 - Salas, Joaquín

PB - Springer Verlag

T2 - 11th Mexican Conference on Pattern Recognition, MCPR 2019

Y2 - 26 June 2019 through 29 June 2019

ER -

Hernández G, Hernández LG, Zamora E , Sossa H, Antelis JM, Mendoza-Montoya O et al. Using Morphological-Linear Neural Network for Upper Limb Movement Intention Recognition from EEG Signals. In Carrasco-Ochoa JA, Martínez-Trinidad JF, Olvera-López JA, Salas J, editors, Pattern Recognition - 11th Mexican Conference, MCPR 2019, Proceedings. Springer Verlag. 2019. p. 389-397. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-030-21077-9_36

Using Morphological-Linear Neural Network for Upper Limb Movement Intention Recognition from EEG Signals

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this