Discreet discrete commands for assistive and neuroprosthetic devices

Document Type

Article

Abstract

Many new assistive devices are available for individuals paralyzed below the neck due to spinal cord injury. Severely paralyzed individuals must be able to command their complex assistive devices using remaining activity from the neck up. Electromyographic (EMG) sensors enable people to use contractions of head and neck muscles to generate multiple proportional command signals. Electroencephalographic (EEG) signals can also be used to generate commands for assistive device control by conveying information about imagined or attempted movements. Fully-implanted wireless biopotential detection systems are now being developed to reliably detect EMGs, EEGs, or a mixture of the two from recording electrodes implanted just under the skin or scalp thus eliminating the need for externally worn hardware on the head or face. This present study shows how novel patterns of jaw muscle contractions, detected via biopotential sensors on the scalp surface or implanted just under the scalp, can be used to generate reliable discrete EMG commands, which can be differentiated from patterns generated during normal activities, such as chewing. These jaw contractions can be detected with sensors already in place to detect other muscle- or brain-based command signals thus adding to the functionality of current device control systems.

Medical Subject Headings

Algorithms; Calibration; Cues; Data Interpretation, Statistical; Discriminant Analysis; Electroencephalography; Electromyography; Humans; Jaw (physiology); Mastication (physiology); Muscle, Skeletal (physiology); Neck Muscles (physiology); Neural Networks, Computer; Prostheses and Implants; Psychomotor Performance (physiology); Reproducibility of Results; Scalp (physiology); Self-Help Devices; Spinal Cord Injuries (rehabilitation)

Publication Date

6-1-2010

Publication Title

IEEE transactions on neural systems and rehabilitation engineering : a publication of the IEEE Engineering in Medicine and Biology Society

E-ISSN

1558-0210

Volume

18

Issue

3

First Page

236

Last Page

44

PubMed ID

20064765

Digital Object Identifier (DOI)

10.1109/TNSRE.2009.2033428

Link to Full Text

Share

COinS