Extracting muscle synergy patterns from EMG data using autoencoders

Martin Spüler; Nerea Irastorza-Landa; Andrea Sarasola-Sanz; Ander Ramos-Murguialday

doi:10.1007/978-3-319-44781-0_6

Extracting muscle synergy patterns from EMG data using autoencoders

Martin Spüler, Nerea Irastorza-Landa, Andrea Sarasola-Sanz, Ander Ramos-Murguialday

Neuroengineering

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

16 Citations (Scopus)

Abstract

Muscle synergies can be seen as fundamental building blocks of motor control. Extracting muscle synergies from EMG data is a widely used method in motor related research. Due to the linear nature of the methods commonly used for extracting muscle synergies, those methods fail to represent agonist-antagonist muscle relationships in the extracted synergies. In this paper, we propose to use a special type of neural networks, called autoencoders, for extracting muscle synergies. Using simulated data and real EMG data, we show that autoencoders, contrary to commonly used methods, allow to capture agonist-antagonist muscle relationships, and that the autoencoder models have a significantly better fit to the data than others methods.

Original language	English
Title of host publication	Artificial Neural Networks and Machine Learning - 25th International Conference on Artificial Neural Networks, ICANN 2016, Proceedings
Editors	Alessandro E.P. Villa, Paolo Masulli, Antonio Javier Pons Rivero
Publisher	Springer Verlag
Pages	47-54
Number of pages	8
ISBN (Print)	9783319447803
DOIs	https://doi.org/10.1007/978-3-319-44781-0_6
Publication status	Published - 2016
Event	25th International Conference on Artificial Neural Networks and Machine Learning, ICANN 2016 - Barcelona, Spain Duration: 6 Sept 2016 → 9 Sept 2016

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	9887 LNCS
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	25th International Conference on Artificial Neural Networks and Machine Learning, ICANN 2016
Country/Territory	Spain
City	Barcelona
Period	6/09/16 → 9/09/16

Keywords

Electromyography (EMG)
Extensor-flexor muscles
Matrix factorization
Neural network

Access to Document

10.1007/978-3-319-44781-0_6

Cite this

Spüler, M., Irastorza-Landa, N., Sarasola-Sanz, A., & Ramos-Murguialday, A. (2016). Extracting muscle synergy patterns from EMG data using autoencoders. In A. E. P. Villa, P. Masulli, & A. J. P. Rivero (Eds.), Artificial Neural Networks and Machine Learning - 25th International Conference on Artificial Neural Networks, ICANN 2016, Proceedings (pp. 47-54). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 9887 LNCS). Springer Verlag. https://doi.org/10.1007/978-3-319-44781-0_6

Spüler, Martin ; Irastorza-Landa, Nerea ; Sarasola-Sanz, Andrea et al. / Extracting muscle synergy patterns from EMG data using autoencoders. Artificial Neural Networks and Machine Learning - 25th International Conference on Artificial Neural Networks, ICANN 2016, Proceedings. editor / Alessandro E.P. Villa ; Paolo Masulli ; Antonio Javier Pons Rivero. Springer Verlag, 2016. pp. 47-54 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{35e86db5a7844efaa03f4320c0d99e20,

title = "Extracting muscle synergy patterns from EMG data using autoencoders",

abstract = "Muscle synergies can be seen as fundamental building blocks of motor control. Extracting muscle synergies from EMG data is a widely used method in motor related research. Due to the linear nature of the methods commonly used for extracting muscle synergies, those methods fail to represent agonist-antagonist muscle relationships in the extracted synergies. In this paper, we propose to use a special type of neural networks, called autoencoders, for extracting muscle synergies. Using simulated data and real EMG data, we show that autoencoders, contrary to commonly used methods, allow to capture agonist-antagonist muscle relationships, and that the autoencoder models have a significantly better fit to the data than others methods.",

keywords = "Electromyography (EMG), Extensor-flexor muscles, Matrix factorization, Neural network",

author = "Martin Sp{\"u}ler and Nerea Irastorza-Landa and Andrea Sarasola-Sanz and Ander Ramos-Murguialday",

note = "Publisher Copyright: {\textcopyright} Springer International Publishing Switzerland 2016.; 25th International Conference on Artificial Neural Networks and Machine Learning, ICANN 2016 ; Conference date: 06-09-2016 Through 09-09-2016",

year = "2016",

doi = "10.1007/978-3-319-44781-0_6",

language = "English",

isbn = "9783319447803",

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

publisher = "Springer Verlag",

pages = "47--54",

editor = "Villa, {Alessandro E.P.} and Paolo Masulli and Rivero, {Antonio Javier Pons}",

booktitle = "Artificial Neural Networks and Machine Learning - 25th International Conference on Artificial Neural Networks, ICANN 2016, Proceedings",

address = "Germany",

}

Spüler, M, Irastorza-Landa, N , Sarasola-Sanz, A & Ramos-Murguialday, A 2016, Extracting muscle synergy patterns from EMG data using autoencoders. in AEP Villa, P Masulli & AJP Rivero (eds), Artificial Neural Networks and Machine Learning - 25th International Conference on Artificial Neural Networks, ICANN 2016, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 9887 LNCS, Springer Verlag, pp. 47-54, 25th International Conference on Artificial Neural Networks and Machine Learning, ICANN 2016, Barcelona, Spain, 6/09/16. https://doi.org/10.1007/978-3-319-44781-0_6

Extracting muscle synergy patterns from EMG data using autoencoders. / Spüler, Martin; Irastorza-Landa, Nerea ; Sarasola-Sanz, Andrea et al.
Artificial Neural Networks and Machine Learning - 25th International Conference on Artificial Neural Networks, ICANN 2016, Proceedings. ed. / Alessandro E.P. Villa; Paolo Masulli; Antonio Javier Pons Rivero. Springer Verlag, 2016. p. 47-54 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 9887 LNCS).

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

TY - GEN

T1 - Extracting muscle synergy patterns from EMG data using autoencoders

AU - Spüler, Martin

AU - Irastorza-Landa, Nerea

AU - Sarasola-Sanz, Andrea

AU - Ramos-Murguialday, Ander

N1 - Publisher Copyright: © Springer International Publishing Switzerland 2016.

PY - 2016

Y1 - 2016

N2 - Muscle synergies can be seen as fundamental building blocks of motor control. Extracting muscle synergies from EMG data is a widely used method in motor related research. Due to the linear nature of the methods commonly used for extracting muscle synergies, those methods fail to represent agonist-antagonist muscle relationships in the extracted synergies. In this paper, we propose to use a special type of neural networks, called autoencoders, for extracting muscle synergies. Using simulated data and real EMG data, we show that autoencoders, contrary to commonly used methods, allow to capture agonist-antagonist muscle relationships, and that the autoencoder models have a significantly better fit to the data than others methods.

AB - Muscle synergies can be seen as fundamental building blocks of motor control. Extracting muscle synergies from EMG data is a widely used method in motor related research. Due to the linear nature of the methods commonly used for extracting muscle synergies, those methods fail to represent agonist-antagonist muscle relationships in the extracted synergies. In this paper, we propose to use a special type of neural networks, called autoencoders, for extracting muscle synergies. Using simulated data and real EMG data, we show that autoencoders, contrary to commonly used methods, allow to capture agonist-antagonist muscle relationships, and that the autoencoder models have a significantly better fit to the data than others methods.

KW - Electromyography (EMG)

KW - Extensor-flexor muscles

KW - Matrix factorization

KW - Neural network

UR - http://www.scopus.com/inward/record.url?scp=84988411363&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-44781-0_6

DO - 10.1007/978-3-319-44781-0_6

M3 - Conference contribution

AN - SCOPUS:84988411363

SN - 9783319447803

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

SP - 47

EP - 54

BT - Artificial Neural Networks and Machine Learning - 25th International Conference on Artificial Neural Networks, ICANN 2016, Proceedings

A2 - Villa, Alessandro E.P.

A2 - Masulli, Paolo

A2 - Rivero, Antonio Javier Pons

PB - Springer Verlag

T2 - 25th International Conference on Artificial Neural Networks and Machine Learning, ICANN 2016

Y2 - 6 September 2016 through 9 September 2016

ER -

Spüler M, Irastorza-Landa N , Sarasola-Sanz A , Ramos-Murguialday A. Extracting muscle synergy patterns from EMG data using autoencoders. In Villa AEP, Masulli P, Rivero AJP, editors, Artificial Neural Networks and Machine Learning - 25th International Conference on Artificial Neural Networks, ICANN 2016, Proceedings. Springer Verlag. 2016. p. 47-54. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-319-44781-0_6

Extracting muscle synergy patterns from EMG data using autoencoders

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this