TY - GEN
T1 - Design of continuous EMG classification approaches towards the control of a robotic exoskeleton in reaching movements
AU - Irastorza-Landa, Nerea
AU - Sarasola-Sanz, Andrea
AU - López-Larraz, Eduardo
AU - Bibián, Carlos
AU - Shiman, Parid
AU - Birbaumer, Niels
AU - Ramos-Murguialday, Ander
N1 - Publisher Copyright:
© 2017 IEEE.
PY - 2017/8/11
Y1 - 2017/8/11
N2 - Myoelectric control of rehabilitation devices engages active recruitment of muscles for motor task accomplishment, which has been proven to be essential in motor rehabilitation. Unfortunately, most electromyographic (EMG) activity-based controls are limited to one single degree-of-freedom (DoF), not permitting multi-joint functional tasks. On the other hand, discrete EMG-triggered approaches fail to provide continuous feedback about muscle recruitment during movement. For such purposes, myoelectric interfaces for continuous recognition of functional movements are necessary. Here we recorded EMG activity using 5 bipolar electrodes placed on the upper-arm in 8 healthy participants while they performed reaching movements in 8 different directions. A pseudo on-line system was developed to continuously predict movement intention and attempted arm direction. We evaluated two hierarchical classification approaches. Movement intention detection triggered different movement direction classifiers (4 or 8 classes) that were trained and tested over a 5-fold cross validation. We also investigated the effect of 3 different window lengths to extract EMG features on classification. We obtained classification accuracies above 70% for both hierarchical approaches. These results highlight the viability of classifying online 8 upper-arm different directions using surface EMG activity of 5 muscles and represent a first step towards an online EMG-based control for rehabilitation devices.
AB - Myoelectric control of rehabilitation devices engages active recruitment of muscles for motor task accomplishment, which has been proven to be essential in motor rehabilitation. Unfortunately, most electromyographic (EMG) activity-based controls are limited to one single degree-of-freedom (DoF), not permitting multi-joint functional tasks. On the other hand, discrete EMG-triggered approaches fail to provide continuous feedback about muscle recruitment during movement. For such purposes, myoelectric interfaces for continuous recognition of functional movements are necessary. Here we recorded EMG activity using 5 bipolar electrodes placed on the upper-arm in 8 healthy participants while they performed reaching movements in 8 different directions. A pseudo on-line system was developed to continuously predict movement intention and attempted arm direction. We evaluated two hierarchical classification approaches. Movement intention detection triggered different movement direction classifiers (4 or 8 classes) that were trained and tested over a 5-fold cross validation. We also investigated the effect of 3 different window lengths to extract EMG features on classification. We obtained classification accuracies above 70% for both hierarchical approaches. These results highlight the viability of classifying online 8 upper-arm different directions using surface EMG activity of 5 muscles and represent a first step towards an online EMG-based control for rehabilitation devices.
UR - http://www.scopus.com/inward/record.url?scp=85034835680&partnerID=8YFLogxK
U2 - 10.1109/ICORR.2017.8009234
DO - 10.1109/ICORR.2017.8009234
M3 - Conference contribution
C2 - 28813806
AN - SCOPUS:85034835680
T3 - IEEE International Conference on Rehabilitation Robotics
SP - 128
EP - 133
BT - 2017 International Conference on Rehabilitation Robotics, ICORR 2017
A2 - Ajoudani, Arash
A2 - Artemiadis, Panagiotis
A2 - Beckerle, Philipp
A2 - Grioli, Giorgio
A2 - Lambercy, Olivier
A2 - Mombaur, Katja
A2 - Novak, Domen
A2 - Rauter, Georg
A2 - Rodriguez Guerrero, Carlos
A2 - Salvietti, Gionata
A2 - Amirabdollahian, Farshid
A2 - Balasubramanian, Sivakumar
A2 - Castellini, Claudio
A2 - Di Pino, Giovanni
A2 - Guo, Zhao
A2 - Hughes, Charmayne
A2 - Iida, Fumiya
A2 - Lenzi, Tommaso
A2 - Ruffaldi, Emanuele
A2 - Sergi, Fabrizio
A2 - Soh, Gim Song
A2 - Caimmi, Marco
A2 - Cappello, Leonardo
A2 - Carloni, Raffaella
A2 - Carlson, Tom
A2 - Casadio, Maura
A2 - Coscia, Martina
A2 - De Santis, Dalia
A2 - Forner-Cordero, Arturo
A2 - Howard, Matthew
A2 - Piovesan, Davide
A2 - Siqueira, Adriano
A2 - Sup, Frank
A2 - Lorenzo, Masia
A2 - Catalano, Manuel Giuseppe
A2 - Lee, Hyunglae
A2 - Menon, Carlo
A2 - Raspopovic, Stanisa
A2 - Rastgaar, Mo
A2 - Ronsse, Renaud
A2 - van Asseldonk, Edwin
A2 - Vanderborght, Bram
A2 - Venkadesan, Madhusudhan
A2 - Bianchi, Matteo
A2 - Braun, David
A2 - Godfrey, Sasha Blue
A2 - Mastrogiovanni, Fulvio
A2 - McDaid, Andrew
A2 - Rossi, Stefano
A2 - Zenzeri, Jacopo
A2 - Formica, Domenico
A2 - Karavas, Nikolaos
A2 - Marchal-Crespo, Laura
A2 - Reed, Kyle B.
A2 - Tagliamonte, Nevio Luigi
A2 - Burdet, Etienne
A2 - Basteris, Angelo
A2 - Campolo, Domenico
A2 - Deshpande, Ashish
A2 - Dubey, Venketesh
A2 - Hussain, Asif
A2 - Sanguineti, Vittorio
A2 - Unal, Ramazan
A2 - Caurin, Glauco Augusto de Paula
A2 - Koike, Yasuharu
A2 - Mazzoleni, Stefano
A2 - Park, Hyung-Soon
A2 - Remy, C. David
A2 - Saint-Bauzel, Ludovic
A2 - Tsagarakis, Nikos
A2 - Veneman, Jan
A2 - Zhang, Wenlong
PB - IEEE Computer Society
T2 - 2017 International Conference on Rehabilitation Robotics, ICORR 2017
Y2 - 17 July 2017 through 20 July 2017
ER -