A new generation of dexterous prosthetic hands is becoming available, which are capable of a greater number of functions than conventional prostheses. To enable this increased functionality, we have pioneered a novel control platform which includes a surface electromyography (EMG) pattern recognition algorithm called MyoSense and a first-of-its-kind multi-channel conformal electrode interface called MyoLiner. The combination of these two technologies has allowed us, for the first time, to give amputees the ability to point their index finger and operate their thumb independently in addition to the conventional hand open and close functionality. This technology has the potential to shift the paradigm in prosthetic control for the first time in over 50 years. In order to maximize the impact and accuracy of the technology, rehabilitation therapy is required, ideally within a 30-day """"""""golden window"""""""" following amputation. However, the realities of patient fitting and medical device reimbursement often delay the introduction of the prosthesis well beyond this period. Based on these considerations, we propose to integrate the innovative MyoLiner and MyoSense technologies with game-based training software into a complete patient-driven rehabilitation system named MyoTrain. Through our partnership with the top ranked video game design program at the University of Southern California, MyoTrain will be specifically designed to: 1) enable patients to practice using myoelectric control as early as possible during the """"""""golden window"""""""", 2) empower individuals to determine their own functional goals , 3) provide meaningful feedback to optimize a patient's ability to control their residual limb muscles, and 4) maximize the patient's engagement through exciting gameplay and meaningful rewards. At the conclusion of this development, we will validate the entire MyoTrain system, after 20 hours of use spread over 8 weeks, on clinically-validated upper limb functionality measures with transradial amputees. The study will be conducted by our partners at the Arm Amputee Program at the National Rehabilitation Hospital in Washington D.C. Upon completion of 20 hours of rehabilitation with the MyoTrain therapy, amputees will be expected to operate at least 6 functions independently with statistically significant improvements in real-time decoding accuracies and on the upper extremity functionality measures. With this improved functionality, we seek to usher in a new era of prosthetic control and improve the effective functionality of dexterous prosthetic devices.
We propose developing novel rehabilitation system that will help amputees strengthen residual limb muscles and regain upper extremity functionality using dexterous myoelectric prostheses.
| Betthauser, Joseph L; Hunt, Christopher L; Osborn, Luke E et al. (2018) Limb Position Tolerant Pattern Recognition for Myoelectric Prosthesis Control with Adaptive Sparse Representations From Extreme Learning. IEEE Trans Biomed Eng 65:770-778 |
| Powell, Michael A; Kaliki, Rahul R; Thakor, Nitish V (2014) User training for pattern recognition-based myoelectric prostheses: improving phantom limb movement consistency and distinguishability. IEEE Trans Neural Syst Rehabil Eng 22:522-32 |
| Powell, Michael A; Thakor, Nitish V (2013) A Training Strategy for Learning Pattern Recognition Control for Myoelectric Prostheses. J Prosthet Orthot 25:30-41 |
