Current Myoprosthetic Developments at the Hugh MacMillan Medical Centre
M. MIFSUD, H. R. GALWAY AND M. MILNER*Toronto, Ontario
Recent myoprosthetic developments at the Hugh MacMillan Medical Centre have been aimed at providing specialized electronic systems to complement the current clinical practice of supplying power-assisted upper-limb myoelectric components for amputees.
The systems described are a miniature Data Acquisition Device (DAD) which collects information about prosthetic use during an amputee's normal activities; a Self-Adaptive Myoelectric Processor (SAMP) that automatically calibrates the system to the myoelectric signal range of an amputee's controlling muscle; a Myoelectric Control Assessment Program (MCAP) which measures objectively, in the clinic, an amputee's ability to operate a prosthesis; and a new device for identifying potential muscle sites for electrode placement. While developed with children in mind, the systems are appropriate for routine use for all ages during training, fitting and subsequent periods.
Data Acquisition Device (DAD)
With the addition of new electromechanical hands, children as young as two or three years are being successfully fitted with below-elbow myoelectric prostheses. As the number of potential users increases and as their age decreases, it becomes more important to have access to reliable and accurate performance measurements regarding. the life expectancy and the efficiency of various prosthetic components.
Until recently, field evaluations of most prostheses have consisted primarily of opinions and observations of the user or, in the case of the young child, comments from the parent. While- these subjective reports are valuable, they do not provide the data needed to evaluate objectively the various operating characteristics of each prosthetic device.
A miniaturized active monitoring circuit, the Data Acquisition Device1,2 was developed in the Hugh MacMillan Medical Centre's Rehabilitation Engineering Department to solve this problem. The DAD is inserted within the prosthesis between the myoelectric control system and the hand (Fig. 1 ). Relying on its own (stand-by) battery power-source, the DAD counts both openings and closings of the hand and measures various time aspects of prosthetic function over a period of up to two months. During the next clinic visit, the DAD is removed and plugged into one of the Centre's microcomputers. The collected data is then computed and tabulated over a 30-second period and a print-out is obtained. After being automatically reset, the DAD can be reinserted into the prosthesis, ready for another two-month period of data collection.
The DAD is being employed successfully to count the total number of open and close activations and the total time the prosthesis is activated during these functions. The microcomputer will use these figures to calculate the ratio of the open-to-close activations and the average time of each activation. By entering into the program the number of user days, as reported by the client, data will also be generated for the average time of daily use and the average number of daily open-and-close activations.
Significant findings to date have indicated that the ratio of activations to time is usually close to 1. Ratios that vary from this are often indicative of myoelectric or prosthesis maladjustment. In addition, prosthesis activation times greater than one hour per day may indicate a heavy user. For such clients, an internally located battery may need to be replaced with an external battery (i.e., Otto Bock). If the average time per activation is longer than one second, the amputee may be contracting the muscle longer than necessary to complete hand opening or closing. Corrective measures could include reducing the sensitivity of the control system or retraining of the user.
Five children whose ages averaged 30 months (range 20-50 months) were provided with myoelectric prostheses for the first time and were monitored with DAD circuits. Their prostheses were equipped with an Otto Bock 2-state system and a Variety Village or Systemteknik child-size hand. Data were collected an average of every 13.625 wearing days.
Recorded results obtained from one child indicated a large number of average activations per day (33142) which were of short duration (.034 second). Improvement in suspension was required. Data collected from the other children indicated an average of 4365 (S.D. = 1981) hand opening and closing activations per day which averaged 0.439 second (S.D. = 0.258). The children used their prostheses actively an average of 38 minutes per day. More information of this kind should help investigations of amputee prosthesis-wearing patterns, identifying problems with prostheses and optimizing battery size for a day's use.
Self-Adaptive Myoelectric Processor (SAMP)
The DAD can also be used to evaluate a recently developed control system called the SAMP or Self-Adaptive Myoelectric Processor3. The SAMP can be used with either 6- or 12-volt components and can be programmed for 2- or 3-state operation.
When a client is restricted to the use of only one muscle for myoelectric control, the variable strength of muscle contraction is used to operate hand opening and closing. (A myoelectric control system developed by the University of New Brunswick is an example of such a control systems). Whereas a weak signal, or the absence of a signal, will generate no hand activity, a slightly stronger signal can be used to close the hand. A progressively stronger contraction will then open the hand. Thus by using these two different levels, or thresholds, a myoelectric device can be made to function with signals from only one muscle. However, much time and effort is required to adjust and set the signal thresholds manually during fitting and training and periodically throughout the lifetime of the prosthesis.
The SAMP was developed as a self-calibration device which would greatly simplify this procedure. After the prosthesis' battery is turned on, the amputee contracts the controlling muscle strongly for at least 5 seconds. The SAMP uses this contraction to set the upper-switching threshold automatically while the lower threshold is set when the muscle is relaxed. The upper threshold will be retained until power to the prosthesis is turned off, while the lower adjusts the threshold automatically with changing "noise" levels in the myoelectric signal.
Use of the SAMP will reduce the number of client visits significantly and facilitate the fitting and the training of young children with 3-state myoelectric control systems.
Ten SAMP circuits (5.56cm.Lx3.73cm.Wx1.53cm.H) have been manufactured for clinical evaluation. They are being used by below-elbow and above-elbow amputees, as well as by a severely disabled individual for accessing microcomputer programs. In general, they have found the system to be useful, allowing more freedom to adjust the system to their daily requirements. Some questions have arisen about poor electrode-skin contact resulting in inaccurate calibration settings. Routine checks of socket fit are indicated for successful implementation.
Myoelectric Control Assessment Program (MCAP)
When assessing the very young client for myoelectric use, the therapist cannot always obtain valid performance levels. Certain complications, such as short and functionally inadequate muscle contractions, muscle contractions which attempt to open or close a hand that has already travelled to its maximum limit, and poor reliability across different evaluators, detract from the objective evaluation of each user. With the development of a microcomputer-assisted, prosthetic assessment device, however, accurate and consistent measurements can be obtained during routine clinic sessions.
The Myoelectric Control Assessment Program (MCAP) enables the client's myoelectric control system to control the movement of a graphic presentation of an electric hand4. The object of each test is to determine the amputee's ability to control the hand's open /close functions. The program also allows the therapist to choose one of three levels of difficulty so as to match each client's level of ability more properly.
Comparison of MCAP evaluations with subjective evaluations by trained personnel has indicated that more reliable measurements are obtained more rapidly using the microcomputer system. Some of the chief advantages of this approach include the microcomputer's ability to store and analyze large amounts of data concerning the open/close control functions (as shown in Chart 1 ), automatically cycle a series of tests, identify correct and incorrect responses and indicate potential problems with the prosthesis or its control system.
MCAP is of potential value in determining if training techniques, prosthesis adjustments or subsequent periods of prosthesis use have affected the amputee's performance measurements. Standardized data collected routinely by those involved in providing upper-limb myoelectric service may also be shared between various clinics to facilitate the identification of amputee performance levels and resolve differences due to the type of control system, amputee's age and level of amputation.
Ready Identification of Muscle Control Site
Additional developments include skin-surface mapping programs to enable the microcomputer to identify potential electrode muscle sites by automatically recording myoelectric activities of various muscles. Two muscles can be mapped simultaneously for comparison of strength, location and isolated muscle control. A typical video monitor display is shown in Figure 2 .
These devices (excluding the SAMP) utilize an inexpensive and readily available Apple 11 plus or lie microcomputer (Fig. 3 ) equipped with a modest number of peripherals.
With increasing use of microcomputers in prosthetic clinics, it is anticipated that the often complicated and limited techniques of muscle-site identification and controls training will soon be replaced by the efficient and inexpensive use of microcomputer technology.
The microcomputer's flexibility for various applications has been described as well as a new control system to enhance the provision of electronic-limb prostheses. The systems have been developed to minimize daily service problems while augmenting traditional procedures with objectively collected data that is easily acquired and standardized so that more informed decisions regarding the amputee's management can be made by clinic personnel. The establishment of pertinent standards relating to the amputee's abilities should facilitate the quick identification of problems and aid in planning patient/clinic schedules. These technological developments should therefore aid both small and large facilities in providing, servicing and maintaining upper-limb myoelectrically controlled prostheses.
*Hugh MacMillan Medical Centre, 350 Rumsey Road, Toronto, Ontario M4G 1 R8, Canada
- Mifsud, M., and M. Milner: A Two Channel Miniature Data Acquisition Device. Accepted for publication in Medical and Biological Engineering and Computing, 1985.
- Mifsud, M., and M. Milner: A Miniature Data Acquisition Device For Amputee Assessment. Accepted for publication in the Proceedings of the Second Canadian Congress of Rehabilitation. Vancouver, 1985.
- Mifsud, M., and M. Milner: Self-Adaptive Myoelectric Processor. Proceedings of IEEE, Electronicom '85, 1985, 2:374-377.
- Mifsud, M., R. Gosine, W. Literowich and M. Milner: A Microcomputer Application for the Measurement of Amputee Performance with Myoelectric Systems. Accepted for publication in the Proceedings of the Second Canadian Congress of Rehabilitation. Vancouver, 1985.
- Richard, P. D.: 3-State Analog Control Unit, in Myoelectric Control Systems, University of New Brunswick, Bio-Engineering Institute, Progress Report #17, 1980, edited by R. N. Scott. 5-10.