Rechargeable Wireless EMG Sensor for Prosthetic Control P.A. Lichter, Member IEEE, E.H. Lange, Member, IEEE, T.H. Riehle, Member, IEEE, S.M. Anderson, Member, IEEE, D.S. Hedin, Member, IEEE
Abstract— Surface electrodes in modern myoelectric prosthetics are often embedded in the prosthesis socket and make contact with the skin. These electrodes detect and amplify muscle action potentials from voluntary contractions of the muscle in the residual limb and are used to control the prosthetic’s movement and function. There are a number of performance-related deficiencies associated with external electrodes including the maintenance of sufficient electromyogram (EMG) signal amplitude, extraneous noise acquisition, and proper electrode interface maintenance that are expected to be improved or eliminated using the proposed implanted sensors. This research seeks to investigate the design components for replacing external electrodes with fullyimplantable myoelectric sensors that include a wireless interface to the prosthetic limbs. This implanted technology will allow prosthetic limb manufacturers to provide products with increased performance, capability, and patient-comfort. The EMG signals from the intramuscular recording electrode are amplified and wirelessly transmitted to a receiver in the prosthetic limb. Power to the implant is maintained using a rechargeable battery and an inductive energy transfer link from the prosthetic. A full experimental system was developed to demonstrate that a wireless biopotential sensor can be designed that meets the requirements of size, power, and performance for implantation.
I. INTRODUCTION
allow for sequential operation of elbow motion, wrist rotation, and hand motions. A myoelectric prosthesis requires EMG signals from select muscles but has the advantage of not needing cables or straps to control the operations of the prosthesis.[2-6] External myoelectric sensors don’t work for all amputees. Often times the electronic gain stages are difficult to adjust and maintain and may not even be a viable option when a patient’s external EMG signal is too weak or scattered. Maintaining a proper electrical interface between the skin and prosthetic control electrode can be difficult. Poor interfacing as well as fluctuations in the signal amplitude from day to day can create performance issues.[1] The proposed technology of using implanted EMG sensors has the potential to reduce or eliminate a number of the current issues of external EMG sensors. Implanted EMG sensors are selective and can be positioned by the orthopedic surgeon to optimize the prosthetic control. Issues of EMG signal amplitude, extraneous noise and proper electrode interface are either eliminated or reduced using implanted sensors. The successful design of in-body wireless transmission requires overcoming unique challenges. Power consumption and size are two important considerations. Both are closely related and must be strongly considered at every stage of design.
O
ver the last few decades, prosthetic control has evolved from conventional body-powered hooks, to movement actuated electric prosthetics. In recent years, the myoelectric prosthesis has evolved to be a preferred alternative to conventional hook prosthesis for patients with traumatic or congenital absence of forearms and hands. In the absence of a hand or arm, the person's brain still continues sending signals to "grasp" or "open" the hand in the residual limb. Surface electrodes often embedded in the prosthesis socket make contact with the skin. These electrodes detect and amplify muscle action potentials from voluntary contractions of the muscle in the residual limb. These amplified electrical signals can control the prosthetic electric motor to provide a prosthetic function, e.g., terminal device operation, wrist rotation, elbow flexion. The newest electronic control systems perform multiple functions, and Manuscript received April 23, 2010. This work was supported in part by the U.S. National Institute of Neurological Disorders and Stroke, National Institutes of Health Grant 1R43NS055424. P.A. Lichter, E.H. Lange, T.H. Riehle, and S.M. Anderson are with Koronis Biomedical Technologies Corp. 6901 E. Fish Lake Road, Suite 190, Maple Grove, MN 55369 USA (phone: (888) 274-1317; e-mail:
[email protected]). D.S. Hedin is with Advanced Medical Electronics, Maple Grove, MN 55369 USA.
II. METHODS A prototype wireless module suitable for a fullyimplanted biopotential sensor to capture signals from intramuscular EMG was designed and built. The aims of the research effort were to demonstrate that a wireless biopotential sensor could be designed to meet the requirements of size, power, and performance required for implantation. The prototype myoelectric interface control for an upper-extremity prosthetic hand needed to meet the key design aspects of a multi-sensor network for prosthetic control (Table I). Table I: Design constraints Requirement
Design Constraints
Data Transmission Rate Power: Battery Voltage Battery Capacity Operating Voltage Transmission Range Circuit Volume Operating Time Antenna Size
20 Kbits/sec 3.8 nominal 10 mAHrs = 20 cm = 16 hours 150µA, Duty Cycle > 20%) does increase the range. The tests were performed using a 5 inch diameter receiver antenna. Transmission Range Tests: Measurements were taken of Range vs. Orientation where the transmitter was in open air and encapsulated within a saline solution. For this testing, data was collected from the receiver onto a PC and the data was inspected for the appropriate bit pattern and determination of Lose-Of-Data (LOD). The transmission signal was inspected at transmission distances of 0, 3, 6, 9, 12, 15, 18 and 21 cm. without any LOD. At each distance tested, the transmitter was evaluated at each of the five locations including four locations on the periphery of the simulated arm each 90 degrees apart and one location site at the center of the arm. EMG Evaluation: The EMG testing included noise floor measurements and comparisons against a commercial EMG test system, the Delsys Bagnoli-2 (see Table II). EMG signals were recorded using surface electrodes at two sites: across the upper arm muscle (Biceps brachii) and across the thumb muscle (Abductor Pollicis Brevis). The EMG signals were recorded simultaneously by each test system and the outputs of the EMG amplifiers were captured on an oscilloscope. Table II: Noise Floor for Prototype and Delsys Systems. System/Configuration Delsys (inputs grounded) Prototype (inputs grounded) Delsys (muscle relaxed) Prototype (muscle relaxed)
RMS 1oise (Refer-To-Input) 1.4 microV 1.6 microV 2.2 microV 2.7 microV
Power Induction Circuit Testing: The power of the excitation circuit measured approximately 65mW and yielded a maximum coil standoff distance of 6.2”. This standoff distance criterion was based upon achieving maximum recharge voltage and current. With the coils positioned at their maximum standoff distance the turn on/off response of the battery charger circuit was captured.
In this study, a wireless myoelectric sensor suitable for application in a prosthetic interface was successfully designed, prototyped and tested. This design was confirmed to meet size and power requirements necessary to be implemented as an implanted neuro-prosthetic sensor. The wireless feasibility testing was directed towards evaluation of transmission range, power consumption and bandwidth. The EMG front-end was evaluated and found to meet requirements necessary for the implant application. ACKNOWLEDGMENT P. A. Lichter thanks William Durfee PhD, Director of Design Education within the Mechanical Engineering Department at the University of Minnesota for his contributions and advisement on this project. REFERENCES [1]
[2]
[3]
[4]
[5]
[6]
[7]
Englehart K., Hudgins B., “A robust, real-time control scheme for multifunction myoelectric control”, Biomedical Engineering, IEEE Transactions on, Volume 50, Issue 7, July 2003 Page(s):848 – 854. Okuno R., Yoshida M., Akazawa K., “Technological developments in Japan – ‘Compliant grasp in a myoelectric hand prosthesis Controlling flexion angle and compliance with electromyogram signals’”, Engineering in Medicine and Biology Magazine, IEEE, Volume 24, Issue 4, July-Aug. 2005 Page(s):48 – 56. Taylor D.M., Tillery S.I.H., Schwartz, A.B., “Information conveyed through brain-control: cursor versus robot”, Neural Systems and Rehabilitation Engineering, IEEE Transactions on, Volume 11, Issue 2, June 2003 Page(s):195 – 199. Fukuda O., Tsuji T., Kaneko M., Otsuka, A., “A human-assisting manipulator teleoperated by EMG signals and arm motions”, Robotics and Automation, IEEE Transactions on, Volume 19, Issue 2, April 2003 Page(s):210 – 222. Goldstein E.A., Heaton J.T., Kobler J.B., Stanley G.B., Hillman R.E., “Design and implementation of a hands-free electrolarynx device controlled by neck strap muscle electromyographic activity”, Biomedical Engineering, IEEE Transactions on, Volume 51, Issue 2, Feb. 2004 Page(s):325 – 332. Routhier F., Vincent C., Morissette M.J., et al., “Clinical results of an investigation of paediatric upper limb myoelectric prosthesis fitting at the Quebec Rehabilitation Institute”. Prosthet Orthot Int. 2001;25(2):119-131. Guoxing W., Wentai L., Bashirullah R., Sivaprakasam M., Kendir G.A., Ying Ji, Humayun, M.S.; Weiland, J.D.; “A closed loop transcutaneous power transfer system for implantable devices with enhanced stability”, Circuits and Systems, 2004. ISCAS '04. Proceedings of the 2004 International Symposium on, Volume 4, 23-26 May 2004 Page(s):IV - 17-20 Vol.4
