EVALUATION OF SHOULDER MOVEMENT AS A COMMAND CONTROL SOURCE

The purpose of this study was to evaluate the use of scapular shoulder movement as a command control source. The focus of this study was on the evaluation of movement signals from quadriplegic subjects as well as on the processing of these signals for use as command inputs to a functional neuromuscular stimulation (FNS) hand-grasp system. The shoulder movement of three CS-level quadriplegic and nine normal subjects was studied using externally mounted two-degree-of-freedom transducers overlying the sternum and clavicles. The C5-level quadriplegic subjects tested had a considerably poorer range of motion than the normal subjects. The range of motion was greatest in elevation but was coupled with a significant component of retraction. The vertical command resolution of the quadriplegic subjects ranged from eight to 13 command levels, whereas the horizontal range contained four or fewer command levels. Normal subjects were able to produce 26 to 79 command levels along the vertical axis and 18 to 85 along the horizontal axis. Subjects were able to maintain a static command level to within several percent of their range in trials lasting from 10 to 30 s. The quadriplegic subjects performed poorly in tasks where they were required to move to a point along the horizontal or vertical central axis and then along the opposite axis. The normal subjects were readily able to perform these tasks. The interference with the command signal due to movement of the opposite extremity was significant, with a maximum error ranging between 25 and 86% along the vertical axis and between 38 and 105% along the horizontal axis. Rise time and normalized velocity (velocity divided by the size of the movement) were found to be suitable for distinguishing different types of shoulder movements independent of the size of the movement. Shoulder movement in quadriplegics is suitable as a command control source for a FNS hand-grasp system requiring one proportional command signal and at least one logical command signal with appropriate processing of the signals. The available transducers are cosmetically acceptable and are easy to don and doff. Use of shoulder movement as a command control source does not interfere with other activities of the user such as eating or talking. Shoulder movement also provides a command control source that is easy for the user to learn and provides some feedback to the user through shoulder proprioception. © 1990 IEEE