Classification of the intention to generate a shoulder versus elbow torque by means of a time-frequency synthesized spatial patterns BCI algorithm

Jie Deng, Jun Yao, Julius P.A. Dewald

Research output: Contribution to journalArticlepeer-review

33 Scopus citations

Abstract

In this paper, we attempt to determine a subject's intention of generating torque at the shoulder or elbow, two neighboring joints, using scalp electroencephalogram signals from 163 electrodes for a brain-computer interface (BCI) application. To achieve this goal, we have applied a time-frequency synthesized spatial patterns (TFSP) BCI algorithm with a presorting procedure. Using this method, we were able to achieve an average recognition rate of 89% in four healthy subjects, which is comparable to the highest rates reported in the literature but now for tasks with much closer spatial representations on the motor cortex. This result demonstrates, for the first time, that the TFSP BCI method can be applied to separate intentions between generating static shoulder versus elbow torque. Furthermore, in this study, the potential application of this BCI algorithm for brain-injured patients was tested in one chronic hemiparetic stroke subject. A recognition rate of 76% was obtained, suggesting that this BCI method can provide a potential control signal for neural prostheses or other movement coordination improving devices for patients following brain injury.

Original languageEnglish (US)
Pages (from-to)131-138
Number of pages8
JournalJournal of neural engineering
Volume2
Issue number4
DOIs
StatePublished - Dec 2005
Externally publishedYes

ASJC Scopus subject areas

  • Biomedical Engineering
  • Cellular and Molecular Neuroscience

Fingerprint

Dive into the research topics of 'Classification of the intention to generate a shoulder versus elbow torque by means of a time-frequency synthesized spatial patterns BCI algorithm'. Together they form a unique fingerprint.

Cite this