One-Muscle Infant's Myoelectric Control
T. WALLEY WILLIAMS, III, MA
A miniature circuit has been developed to permit one-muscle voluntary-opening control of electric hands. This circuit has been fitted to more than 50 persons, mostly infants between 7 months and 3 years. Little training is required to learn to operate a hand with the circuit when triggered by a typical myoelectrode amplifier. Therapy is used to bring the infant to a level of function that is age appropriate.
The Variety Ability Systems 0-3 hand now makes it possible to fit infants at 7 months of age. It should not be surprising that the optimal control for such a young child might differ from the best adult control system.
Over the last three years, we have worked with a number of prosthetists to achieve the simplest possible control system. The process started when Thomas P. Haslam, CP of Medical Center Prosthetics in Houston asked for a single-muscle, voluntary-opening myoelectric control for use with small children treated at the Saint Anthony Center. The first circuit was built to control a Systemteknik 2-6 hand. To understand the rationale behind the new circuit, one should know how conventional systems work.
How Electric Hands Work
From the engineer's viewpoint, all so-called myoelectric hands are, in reality, electric hands controlled by digital logic signals on two control wires. The hands themselves are supplied by two additional power wires from a 6V battery. Thus there are four lines, two power and two control. For convenience, the negative side of the battery is the reference level. When both control lines are at the reference potential or 0V, the hand remains motionless. If a logic one (a positive voltage of about 5V) is applied to the open-hand line, the motor drives the fingers apart. A positive signal on the close-hand line will close the fingers. Two positive signals are, of course, possible; and the hand control circuit must cope by either keeping the hand stationary or by giving preference to the first signal to reach the triggering voltage. Both schemes are used by hand manufacturers.
Conventional Myoelectric Control
For twenty years, people have controlled electric hands by supplying open and close signals from muscle tension detectors placed over the wrist flexors and extensors or some other agonist-antagonist pair. The most popular detector worldwide is the Otto Bock 13E67 electrode assembly. For patients with only a single suitable muscle, the University of New Brunswick Bio-Engineering Institute has developed circuits which generate a close signal for small muscle contractions and an open signal for large. This scheme has been widely used with children. Contrary to popular opinion, the system is easily mastered by children and is often prescribed for those who could just as well use a two-muscle system. It is preferred because there are no electrode bulges. Recently, Otto Bock introduced an electrode which requires a special hand control circuit. It responds to the speed of contraction of a single muscle. The special hands are only available for adults at the present time.
Stalled Motor Problem
As an electric motor slows, it draws more current. Thus a well-designed, battery-operated device should avoid having the motor stalled for longer than a few milliseconds. This can be achieved by use of a slip clutch (Otto Bock) or by a timing circuit (Steeper and Variety Ability Systems) or by ignoring the problem and wasting power (Systemteknik). The problem with Systemteknik hands is sufficiently bad that the University of New Brunswick developed a battery saver circuit to be connected between the control lines and the hand circuit. It introduces a timer that limits the close-hand signal to just the time required to close the hand fully. The circuit must detect an open signal before it will reset. Another way to fix a Systemteknik hand is the substitution of an electronics pack with a built-in timer designed by Variety Ability Systems. Liberty Mutual has made both options available to those fitting Systemteknik hands.
Simple Voluntary-Opening Control
The simplest one-muscle control circuit generates a positive voltage on the close line at all times except when the circuit receives a signal from the muscle. Then it generates a positive signal on the open line. This simple circuit presumes that the controlled device has a timer or other battery saver feature for both the open and close functions. Otherwise the hand will draw power continuously except when open.
Size Problem with Circuits for Infants
There is little room inside an infant prosthesis. Thus, unnecessary connectors should be avoided. The circuit designed by the author for Liberty Mutual is a module measuring only 1/8 x 1/2 x I inch (10 x 13 x 25 mm) with all of the appropriate wires exiting without connectors. This is smaller than just the connectors on most circuits. Our most popular infant control has a receptacle for an Otto Bock 757B8 Battery on the end of a child-proof cable 22 inches (560 mm) long, a connector for a Bock 13E67 electrode assembly on a 4 inch (110 mm) cable, and a Bock four socket connector on a short 2 inch (50 mm) cable for the hand. The circuit is offered in any other variation that might prove useful as shown in Table 1 .
Cookie Crusher Problem
We refer to our circuit as "the cookie crusher circuit" so that everyone will be aware that the hand has only one speed when it closes-fun force. This force is not a problem with the Variety Ability Systems V03 hand, but is with the Systemteknik hand which has a 10-pound (4.5N) pinch. Both parents and child must understand the implications of a 10-pound pinch before this combination is prescribed.
New Bock 13E125 Electrode
Otto Bock has recently introduced a smaller electrode with a variable voltage output to replace its digital output electrode. The old digital electrode is adjusted to give a logic one (positive voltage) output when the muscle tension reaches the threshold set on a dial on the back. The output of the new electrode is controlled by a similar dial, but there is no threshold, just a voltage that varies between zero and about 4 volts. We get around the "no threshold" problem by building a threshold into our circuit. We have set this threshold so that the settings on the new electrode will correspond to the output needed to trigger the circuit with both weak and strong muscle signals. (See SA501 and SA502 in Table 1 and Table 2 )
Training with the Cookie Crusher Circuit
It is a mistake to assume that because this new circuit makes operation easy, no training or therapy is required. Infants require a reward when they contract the correct muscle to open the hand. Several institutions have used a carousel mounted over a crib and set to operate only when the electrode senses a contraction. For the first session the electrode can be held in place by the therapist. Thereafter it is mounted in a temporary socket. Therapy should continue after the prosthesis is provided until the infant is doing age-appropriate activities with the new hand.
Benefits of Early Fitting
"Because we can do it, we ought to." This statement characterizes a lot of the pall mall way we advance human technology. We can fit infants at seven months. We have clear anecdotal evidence that by 18 months these children have excellent bilateral function. Does this mean that we should fit all children at this early age with a myoelectric prosthesis if it is technically possible? I urge you to ask how we can answer this question in a reasonably quantitative way. For instance, should we administer a standardized test to 100 children, fit between 7 and 12 months when these children reach 4 to 41/2 years of age, and compare their function with a control group treated more conventionally?
Liberty Mutual Research Center, 71 Frankland Road, Hopkinton, MA 01748