No abstract

Pattern recognition–based control of myoelectric prostheses has shown great promise in research environments, but has not been optimized for use in a clinical setting. To explore the relationship between classification error, controller delay, and real-time controllability, 13 able-bodied subjects were trained to operate a virtual upper-limb prosthesis using pattern recognition of electromyogram (EMG) signals. Classification error and controller delay were varied by training different classifiers with a variety of analysis window lengths ranging from 50 to 550 ms and either two or four EMG input channels. Offline analysis showed that classification error decreased with longer window lengths (p < 0.01). Real-time controllability was evaluated with the Target Achievement Control (TAC) Test, which prompted users to maneuver the virtual prosthesis into various target postures. The results indicated that user performance improved with lower classification error (p<0.01) and was reduced with longer controller delay (p<0.01), as determined by the window length. Therefore, both of these effects should be considered when choosing a window length; it may be beneficial to increase the window length if this results in a reduced classification error, despite the corresponding increase in controller delay. For the system employed in this study, the optimal window length was found to be between 150 and 250 ms, which is within acceptable controller delays for conventional multi-state amplitude controllers.

Abstract-Despite high classification accuracies (~95%) of myoelectric control systems based on pattern recognition, how well offline measures translate to real-time closed-loop control is unclear. Recently, a real-time virtual test analyzed how well subjects completed arm motions using a multiple-degree of freedom (DOF) classifier. Although this test provided real-time performance metrics, the required task was oversimplified: motion speeds were normalized and unintended movements were ignored. We included these considerations in a new, more challenging virtual test called the Target Achievement Control Test (TAC Test). Five subjects with transradial amputation attempted to move a virtual arm into a target posture using myoelectric pattern recognition, performing the test with various classifier (1-vs 3-DOF) and task complexities (one vs three required motions per posture). We found no significant difference in classification accuracy between the 1-and 3-DOF classifiers (97.2% +/-2.0% and 94.1% +/-3.1%, respectively; p = 0.14). Subjects completed 31% fewer trials in significantly more time using the 3-DOF classifier and took 3.6 +/-0.8 times longer to reach a three-motion posture compared with a onemotion posture. These results highlight the need for closed-loop performance measures and demonstrate that the TAC Test is a useful and more challenging tool to test real-time pattern-recognition performance.

Amputees cannot feel what they touch with their artificial hands, which severely limits usefulness of those hands. We have developed a technique that transfers remaining arm nerves to residual chest muscles after an amputation. This technique allows some sensory nerves from the amputated limb to reinnervate overlying chest skin. When this reinnervated skin is touched, the amputees perceive that they are being touched on their missing limb. We found that touch thresholds of the reinnervated chest skin fall within near-normal ranges, indicating the regeneration of largefiber afferents. The perceptual identity of the limb and chest was maintained separately even though they shared a common skin surface. A cutaneous expression of proprioception also occurred in one reinnervated individual. Experiments with peltier temperature probes and surface electrical stimulation of the reinnervated skin indicate the regeneration of small diameter temperature and pain afferents. The perception of an amputated limb arising from stimulation of reinnervated chest skin may allow useful sensory feedback from prosthetic devices and provides insight into the mechanisms of neural plasticity and peripheral regeneration in humans.regeneration ͉ neural machine interface ͉ touch ͉ artificial limbs T he loss of an arm is a singularly debilitating injury. Improving the function of artificial arms remains a considerable challenge, especially for high-level amputations where the disability is greatest. A primary impediment to better function is that current prostheses provide very little sensory feedback. The amputees must rely primarily on vision to manipulate objects, and they cannot feel what they touch with motorized prosthetic hands. This limitation greatly increases the cognitive burden on the amputee and impedes the use of the artificial limb.We have developed a neural-machine interface called targeted reinnervation (TR) that provides enhanced motor control and the potential for meaningful sensation feedback for artificial arms (1-3). The amputated brachial plexus nerves that once provided motor control and sensory feedback in the missing limb are transferred to arm and chest muscles that remain after the amputation. Once reinnervated, these muscles produce electromyogram (EMG) signals that correspond to the original arm control signals sent from the brain down the brachial plexus nerves. This process provides improved and more intuitive control of a motorized artificial arm (4-6). The contractions of the reinnervated muscles function as biological amplifiers for the motor commands transmitted by the amputated arm nerves. Concurrently, the sensory nerve fibers in these amputated nerves appear to reinnervate the skin overlying the target muscles. When this reinnervated skin is touched, the amputee feels as if the missing hand is being touched. This skin reinnervation may provide a direct portal to the sensory pathways of the amputated arm and hand. It could potentially provide an amputee with the ability to feel what he touches with a pro...

Pattern Recognition is a useful tool for deciphering movement intent from myoelectric signals. Recognition paradigms must adapt with the user in order to be clinically viable over time. Most existing paradigms are static, although two forms of adaptation have received limited attention. Supervised adaptation can achieve high accuracy since the intended class is known, but at the cost of repeated cumbersome training sessions. Unsupervised adaptation attempts to achieve high accuracy without knowledge of the intended class, thus achieving adaptation that is not cumbersome to the user, but at the cost of reduced accuracy. This study reports a novel adaptive experiment on eight subjects that allowed repeated measures post-hoc comparison of four supervised and three unsupervised adaptation paradigms. All supervised adaptation paradigms reduced error over time by at least 26% compared to the nonadapting classifier. Most unsupervised adaptation paradigms provided smaller reductions in error, due to frequent uncertainty of the correct class. One method that selected high-confidence samples showed the most practical implementation, although the other methods warrant future investigation. Supervised adaptation should be considered for incorporation into any clinically viable pattern recognition controller, and unsupervised adaptation should receive renewed interest in order to provide transparent adaptation.