Link to Figures Show Figures
	Article Text Figures References
	Send Link Send HTML Article

Myoelectric Hand Orthosis

Nisim Benjuya, Ph.D.
Steven B. Kenney, B.S.M.E.

Abstract

A new myoelectric hand orthosis, primarily for the spinal cord injury patients at the C5-6 level, is developed. The device, weighing only 500 grams, operates proportionally to the bioelectric potentials picked-up by the surface electrodes. The device offers compatibility with the daily activities of the user without interference and can be donned and doffed independently.

Introduction

Researchers in rehabilitation engineering design orthoses to compensate or substitute for an impaired organism by offering a more normal or an alternate mode of function. A major function of the hand orthosis (HO) is to provide the ability to pinch, grip, and release objects during daily living activities. The ideal HO uses the skeletal structure and the biomechanical properties of the users limb as an integral part of its powered mechanical system.

Over the past two decades, investigators have built devices that were artificially powered or body powered^1-9 The tendency of powered devices was and is to stress greater sophistication of the control unit. In the early 1960s, after electromyography (EMG) proved successful, researchers adapted EMG signals for control with prostheses. One of the most significant advantages of EMG control signals is that they can be obtained from the action potentials of contracting muscle fibers despite little or no movement, making the device attractive for patients with weak or small muscles. EMG, unlike other control signals, carries the distinct signature of the voluntary intent of the central nervous system (CNS). Researchers have developed techniques to extract the wealth of information available in the EMG signal, so as to properly interpret the EMG signal for purposeful activation of such devices as neuromuscular stimulators.^10,11 These recent advances overshadow some earlier reports that recommended a search for alternative signal sources other than EMG.¹²

New surface electrodes that have high selectivity and smart microprocessors can discriminate between various signal characteristics in a proportional manner. Researchers have implemented them in newly designed upper extremity orthotic and prosthetic devices.^13-16 These new devices offer effortless operation with satisfactory results. However, complexity of use, reliability of control, cosmesis and level of independence provided are factors that influence how well potential users accept these devices.

One of the most popular non-powered devices for prehension, the wrist driven flexorhinge tenodesis splint, has received attention both in clinical settings and at the engineering bench. Therapists routinely order bilateral, wrist-driven flexor-hinge splints early in rehabilitation programs. Such splints provide prehension and strengthen wrist extension. However, for reasons of cosmesis and lack of pinch force, the primary use of these splints' remains limited to wrist strengthening.

We designed an HO for spinal cord injury (SCI) patients at the C5,6 level primarily, with a chronic, functional deficit of the hand and wrist. This target population is one of the most challenging ones when it comes to acceptance of an HO or of any other external device. Some design considerations for this population that numerous investigators have faced over the last two decades include: functional efficiency, weight, simplicity, expense, durability, and cosmetic appeal. However, with the increasing number of SCI patients due to better patient care and the limited number of aides to the disabled, a need has arisen for different and more demanding orthotic systems. Factors considered in the design of the myoelectric powered HO reported in this paper include: (1) independence in donning and doffing; (2) independence in operation; (3) successful integration with other activities requiring hand manipulation; (4) a control that lets the patient perform his activities as naturally as possible and with minimum conscious effort; and (5) visual feedback provided to the user.

The aim of this research effort was to evaluate the increase in voluntary function of the hand. The investigators evaluated the effects of the orthosis when worn during daily activity and dynamic hand movement.

Description of the Orthosis

The HO (Figure 1) consists of two parts: a hand piece and a forearm piece that are spanned by a flexible shaft. The hand piece is made from one of several molds of various sizes of normal hands, and is easily reshaped by heat for individual adaptation. We designed the hand piece to support and realign the metacarpal arch and stabilize the thumb in opposition to the index and middle fingers. A fingerpiece attached to the gearbox on the hand piece guides these two fingers. The gearbox consists of a spur gear aligned with the metacarpophalangeal joint of the index finger and of a worm gear fixed on the hand piece above the thumb (Figure 2) . The design considerations were adopted after careful study of hand use during various functions.

The forearm piece houses the device's hardware, including the surface EMG electrodes, a miniature DC motor, a rechargeable battery and a control circuit. The first design consideration for the forearm piece was that the EMG electrodes automatically align with the target muscle(s), that is, with the wrist extensors and in some cases the wrist flexors (target muscles vary with the patient). Via a hook sewn to the forearm piece, the patient can pull the forearm band over the limb and position it relative to the muscle(s). The patient then turns on the HO and activates it at will. The second design consideration for the forearm piece was cosmetic: to hide as many components as possible under the patient's long sleeve shirt.

The miniature DC motor (Portescap M16.16) is controlled by elbow myoelectrodes and control circuitry created in cooperation with Liberty Mutual of Boston. The motor control is in a proportional manner to the EMG amplitude based on pulse width (PW) modulation. The electrodes consist of two steel tips and were designed for adequate contract with the control site. The motor is fixed on the forearm piece posteriorly in line with the transmission unit mounted on the hand piece. The flexible shaft interconnects the shafts of the DC motor and the worm gear, energizing the finger to close at a pinch force of 6-7 lbs. and open as much as 10 cm within 5.5-6 seconds. The pinch force is displayed by the 5 LED array mounted on the cover of the transmission box. LED light, in a logarithmic order, covers the range of pinch up to 5 lbs. The smooth switching of LEDs lets the user distinguish mid ranges of pinch force.

The hand and forearm pieces combined weigh approximately 500 grams. A rechargeable 6V nickel-cadmium battery by OttoBock accounts for most of the weight. Under normal operation, this battery provides eight to 12 hours of device use.

Principles of Myoelectric Control

The orthosis (Figure 3) offers two versions of control that cater to a variety of pathologies: one option has one EMG site-electrode and the second option has two EMG sites-electrodes. The first option is based on two degrees of freedom control: when the muscle contracts the hand closes, the speed of movement being proportional to the intensity of the myoelectric signal; the absence of a signal opens the hand. A brief contraction of motivating muscle halts the hand at the desired open hand position. The second option offers three degrees of freedom control. The EMG electrode #1 sends either a closing or opening signal, and the EMG electrode #2 sends the opposite signal to that of #1; no muscular contraction maintains the position reached by the fingers. This option lets the patient sustain the pinch force without any muscular contraction, unlike the first option that requires sustained myoelectric activity during hold and the hand can only open all the way. In both options, the pinching power is the same and proportional to the muscular contraction.

Clinical Trials

The one EMG-site version has been tried on two C6 (complete) SC! patients and one brachial plexus (BP) patient. Both SC! patients have deltoid and biceps of the upper extremity intact with sufficient residual wrist extensor activity (of the extensor carpi radialis longus and brevis) to allow myoelectric signal detection.

The SCI patients required as little as 10 minutes of training for activation of the HO with minimal conscious effort. In order to integrate use of the orthosis with a more immediate need of the patient, that is maneuvering the wheelchair, push (friction) gloves were worn over the hand piece (Figure 4) . The HO was turned off when the patient propelled the wheel chair. Daily activities such as feeding, shaving and teeth brushing were accomplished, but required that the patient be more familiar with using HO. Both patients are currently using the HO with periodic follow-up.

The BP patient required a slightly different fitting since the EMG site wasn't in the forearm band range. The pectoralis major was selected as the EMG site because it interfered the least with the patients' other daily activities, thereby minimizing accidental activation of the device. This patient continues to wear the HO sporadically together with a Roehampton flail arm splint.

Another BP patient favored the two EMG site version, with electrodes positioned on two natural antagonists (biceps/triceps) of the intact arm. This patient is undergoing fitting and training of the flail arm splint in order to use the HO effectively.

Both the SCI and BP patients are enthusiastic about the HO and have challenged our group to develop and respond to their individual needs.

Acknowledgements

The authors wish to acknowledge the contributions of Elliot Lach, M.D., Steve Greelish, and Vincent Durso, C.O., to this report.

This work was supported by the Veterans Administration RR&D Service.

Nisim Benjuya, Ph.D., is project director at Rehabilitation Engineering R&D Lab, VA Medical Center, W. Roxhury, Massachusetts 02132.

Steven B. Kenney, BSME, is currently attending the Graduate School of the University of Arizona.

References:

1. Antonelli, D.J. and W. Worden, "Myoelectric Control of Powered Devices," Arch. Phys. Med. Rehab., 7, 1967, pp.345-349.
2. Engen, T.J. and W.A. Spencer, "Development of Externally Powered Upper Extremity Orthotics," Texas Institute for Rehabilitation and Research, Houston, 1969.
3. Janovsky, F.A., "Eine myoelektrisch gesteuerte Orthese zur Versorgung von Ouerschnitts-Gelahmten," Orthopadie Technik, 9,1972, pp. 261-265.
4. Schuy, S.. G. Wiebpeiner and R. Myoelectrically Controlled Externally Powered Orthosis for Upper Extremities," Biomed. Technik, 26, 1981, pp. 291-296.
5. Hamonet, C., A. de Montgolfier, M. Bedoiseau, P. Lacert, S. Pannier, J.C. Raoult, F. Fournier, G. Carre, D. Meziere, and A. Grossiord. "Myoelectrically Controlled Orthoses for Paralytic Hand," Ann. Chir., 27, 1973, pp. 541-545.
6. Watanabe, H., K. Ogata, T. Okabe, and T. Amano, "Hand Orthosis for Various Finger Impairments ä The K U finger Splint," Prosthetics and Orthotics International, 2,1978, pp. 95-100.
7. Simpson, D.C., "The Choice of Control System for the Multimovement Prosthesis, Extended Physiological Proprioception (E.P.P.)," Proceedings of International Symposium on the Control of Upper Extremity Prosthesis and Orthoses, Goteborg, Sweden, 1971.
8. Le Blanc, M.A., "Innovation and Improvement of Body Powered Arm Prostheses: A First Step," Clinical Prosthetics and Orthotics, Winter 1985.
9. Engel, W.H., M.A. Kmiotek, J.P. French, Mi. Barnerias, and A.A. Siebens, "A Functional Splint for Grasp Driven by Wrist Extension," Arch. Phys. Med., 48, 1967, pp. 43-52.
10. Graupe, D., and W.K. Cline, "Functional Separation of EMG Signals via ARMA Identification Methods for Prosthesis Control Purposes," IEEE Trans. Syst. Man Cybern., 5(2), 1975, p. 252.
11. Graupe, D., J. Salahi, and K.H. Kohn, "Multifunctional Prosthesis and Orthosis Control via Microcomputer Identification of Temporal Pattern Differences in Single-Site Myoelectric Signals," I. Biomed. Eng., 14, 1982, pp. 17.
12. Long II, C., "Normal and Abnormal Control in the Upper Extremities," Final Report, Social & Rehabilitation Services, Grant No. RD-2377M, April 1970.
13. A.B. Cameron, and V.A. Dunfield, "Sensory Feedback System Compatible with Myoelectric Control," Med. Biol. Eng. Comput., 18, 1980, 65-69.
14. Mann, R.W., and S.D. Reimers, "Kinesthetic Sensing for the EMG Controlled `Boston Arm, IEEE Trans. Man-Mach. Syst., MMS-11, 1, 1970, pp. 110-115.
15. Childress, D.S., D.W. Holmes, and J.N. Billock, "Ideas on Myoelectric Prosthetic Systems for Upper-Extremity Amputees," The Control of Upper Extremity Prostheses and Orthoses, 1971, 86-106.
16. Stein, R.B., D. Charles, JA. Hoffer, J. Arsenault, L.A. Davis, S. Moorman, and B. Moss, "New Approaches for the Control of Powered Prostheses, Particularly by High-Level Amputees," Bull. Prosth. Res., BPR 10-33, 17, Spring 1980, pp. 51-62.