Performance of pattern recognition myoelectric control using a generic electrode grid with Targeted Muscle Reinnervation patients

Abstract

Targeted muscle reinnervation (TMR) is a surgical technique that creates myoelectric prosthesis control sites for high-level amputees. The electromyographic signal patterns provided by the reinnervated muscles are well-suited for pattern recognition (PR) control. PR control uses more electrodes compared to conventional amplitude control techniques but their placement on the residual limb is less critical than for conventional amplitude control. In this contribution, we demonstrate that classification error and real-time control performances using a generically placed electrode grid were equivalent or superior to the performance when using targeted electrode placements on two transhumeral amputee subjects with TMR. When using a grid electrode layout, subjects were able to complete actions 0.290 sec to 1 sec faster and with greater accuracy as compared to clinically localized electrode placement (mean classification error of 1.35% and 3.2%, respectively, for a 5 movement-class classifier).These findings indicate that a grid electrode arrangement has the potential to improve control of a myoelectric prosthesis while reducing the time and effort associated with fitting the prosthesis due to clinical localization of control sites on amputee patients.