Event Abstract

BCI Optimization for 3D Continuos Hand Movement Reconstruction from EEG and EMG recordings

  • 1 Graduate School of Engineering, Chiba University, Japan
  • 2 Center for Medical Engineering, Chiba University, Japan

In the field of movement reconstruction, hand and foot prostheses have reached high accuracy[1], [2]. However, success has been limited to cases where the task that the prosthesis performs had a finite number of states. In these cases, most of the limb was still present, and therefore, relevant muscle activity still remained. Thus, it was possible to collect relevant information related with specific movements through electromyography (EMG). In the case of many upper limb amputees, such as persons with trans-humeral amputation, typically only a few muscles directly related to upper limb movement remain. Since available information from the muscles is reduced, Brain Computer Interfaces (BCIs) are used to increase the movement reconstruction accuracy[3]–[5]. Studies using Electroencephalography (EEG) for predicting three dimensions reach low accuracy (around 0.5 correlation between real movement and reconstructed movement)[6]. In this study we used a real world applicable approach to the problem using both EEG and EMG information. We decided to use both EEG and EMG because the EEG is too noisy to get precise information of upper limb movements. Nevertheless, it contains a variety of information which could be used to improve the system. On the other hand, EMG signals have a high correlation with the movement, but this information is very local. Therefore, EMG around the shoulder does not contain as precise information about the hand’s movement as forearm EMG would have. By combining EMG and EEG recordings we expect to improve the general information extracted from the EEG with the local information from the EMG signals. To the best of our knowledge, there is no study using this method for 3D hand movement reconstruction. In this study we propose four different predictors based on ANN to solve the problem (see Figure 1). Moreover, we use EEG and EMG independently to make clear their role. Furthermoer we optimize one of the predictors and analyze the importance of each one of the channels. In this study with a data set collected from 16 subjects, we obtained high correlation (coefficients up to 0.891) between recorded and reconstructed hand movement. Figure 2 shows the comparison between the original movement (red) and the predicted one (blue) for three dimensions x(bot), y (mid) and z(top). This particular subject has a correlation value of 0.804.

Figure 1
Figure 2


We would like to thank his help with the proof reading and corrections to Pablo Varona and for the graphic composition to Jorge Femenía.


[1] F. Tenore, A. Ramos, A. Fahmy, S. Acharya, R. Etienne-Cummings, and N. V Thakor, “Towards the control of individual fingers of a prosthetic hand using surface EMG signals.,” Conf. Proc. ... Annu. Int. Conf. IEEE Eng. Med. Biol. Soc. IEEE Eng. Med. Biol. Soc. Annu. Conf., vol. 2007, pp. 6146–9, Jan. 2007.
[2] S. Au, M. Berniker, and H. Herr, “Powered ankle-foot prosthesis to assist level-ground and stair-descent gaits.,” Neural Netw., vol. 21, no. 4, pp. 654–66, May 2008.
[3] J. Lv, Y. Li, and Z. Gu, “Decoding hand movement velocity from electroencephalogram signals during a drawing task.,” Biomed. Eng. Online, vol. 9, p. 64, 2010.
[4] J.-H. Kim, F. Bieβmann, S.-W. Lee, F. BieBmann, and S.-W. Lee, “Reconstruction of hand movements from EEG signals based on non-linear regression,” in 2014 International Winter Workshop on Brain-Computer Interface (BCI), 2014, pp. 1–3.
[5] N. Robinson, A. P. Vinod, and C. Guan, “Hand movement trajectory reconstruction from EEG for brain-computer interface systems,” in Proceedings - 2013 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2013, 2013, pp. 3127–3132.
[6] T. J. Bradberry, R. J. Gentili, and J. L. Contreras-Vidal, “Reconstructing three-dimensional hand movements from noninvasive electroencephalographic signals.,” J. Neurosci., 2010.

Keywords: BCI, Movement Reconstruction, EEG, EMG, prosthetic control

Conference: German-Japanese Adaptive BCI Workshop, Kyoto, Japan, 28 Oct - 29 Oct, 2015.

Presentation Type: Poster presentation

Topic: Adaptive BCI

Citation: Fernandez-Vargas J, Chu L, Kita K and Yu W (2015). BCI Optimization for 3D Continuos Hand Movement Reconstruction from EEG and EMG recordings. Front. Comput. Neurosci. Conference Abstract: German-Japanese Adaptive BCI Workshop. doi: 10.3389/conf.fncom.2015.56.00021

Copyright: The abstracts in this collection have not been subject to any Frontiers peer review or checks, and are not endorsed by Frontiers. They are made available through the Frontiers publishing platform as a service to conference organizers and presenters.

The copyright in the individual abstracts is owned by the author of each abstract or his/her employer unless otherwise stated.

Each abstract, as well as the collection of abstracts, are published under a Creative Commons CC-BY 4.0 (attribution) licence (https://creativecommons.org/licenses/by/4.0/) and may thus be reproduced, translated, adapted and be the subject of derivative works provided the authors and Frontiers are attributed.

For Frontiers’ terms and conditions please see https://www.frontiersin.org/legal/terms-and-conditions.

Received: 14 Sep 2015; Published Online: 04 Nov 2015.

* Correspondence: Mr. Jacobo Fernandez-Vargas, Graduate School of Engineering, Chiba University, Chiba, Japan, jacobofv@chiba-u.jp

© 2007 - 2019 Frontiers Media S.A. All Rights Reserved