Frontiers in Neuroscience (Dec 2014)

Brain-Machine Interface to Control a Prosthetic Arm with Monkey ECoGs during Periodic Movements

  • Soichiro eMorishita,
  • Keita eSato,
  • Hidenori eWatanabe,
  • Yukio eNishimura,
  • Tadashi eIsa,
  • Ryu eKato,
  • Tatsuhiro eNakamura,
  • Hiroshi eYokoi

DOI
https://doi.org/10.3389/fnins.2014.00417
Journal volume & issue
Vol. 8

Abstract

Read online

Brain Machine Interfaces (BMIs) are promising technologies to rehabilitate the function of upper limbs in severely paralyzed patients. We succeeded in developing a BMI prosthetic arm for a monkey implanted with electrocorticogram (ECoG) electrodes and trained in a reaching task. It had stability in preventing the misclassification of ECoG patterns. However, the latency was about 200 ms as a trade-off for the stability. To improve the response of this BMI prosthetic arm, the generation of a trigger event by decoding muscle activity was adopted. It was performed to predict integrated electromyograms (iEMGs) from the ECoGs. Experiments were conducted to verify the availability of this method, and the results confirmed that the proposed method was superior to the conventional one. In addition, a performance test of the proposed method with actually achieved iEMGs instead of predicted iEMGs was performed, and we found that the motor intention is finely expressed through estimated muscle activity from brain activity rather than actual muscle activity.

Keywords