Simulated feedforward neural network coordination of hand grasp and wrist angle in a neuroprosthesis.

This study presents a possible solution of the general problem of coordinating muscle stimulation in a neuroprosthesis when multiarticular muscles introduce mechanical coupling between joints. In a hand-grasp neuroprosthesis, extrinsic hand muscles cross the wrist joint and introduce large wrist flexion moments during grasp. In order to control hand grasp and wrist angle independently, a controller must take the mechanical coupling into account. In simulation, we investigated the use of artificial neural networks to coordinate hand and wrist muscle stimulation. The networks were trained with data that is easily obtained experimentally. Feedforward control showed excellent hand and wrist coordination when the properties of the system were fixed and there were known external loads. Predictable disturbances (e.g., gravity acting on the hand) can be compensated by sensing arm orientation. However, since wrist angle is sensitive to unpredictable disturbances (e.g., fatigue or object weight), voluntary intervention or feedback control may be required to reduce residual errors.

Biaxial strain properties of elastase-digested porcine aortic valves.

BACKGROUND AND AIMS OF THE STUDY: Previous studies have suggested that elastin in porcine aortic valve cusps is responsible for restoring collagen fibers to their original configuration between loading-unloading cycles. METHODS: Biaxial loading tests were performed on intact aortic valves before and after elastase treatment to further investigate the role of elastin. RESULTS: Degradation of elastin caused an increase in the radial dimensions of the cusps (mean increase in gauge length, 29%), which corresponded to a significant decrease in radial extensibility (mean decrease, 61%) and a threefold increase in radial stiffness. Changes in circumferential extensibility and stiffness were smaller and, for most cusps, were not statistically significant. Control experiments, in which the valves were treated with buffer only, resulted in the opposite changes in radial dimensions and extensibility (7% decrease in gauge length and doubling of extensibility). CONCLUSION: Changes in the mechanical properties of the aortic valve cusps following incubation in elastase were due to elastin damage, and not incidental to soaking in buffer. As many explanted bioprosthetic valves have mechanical characteristics similar to those of the elastase-treated valves, elastin damage may be a factor in the progressive degeneration and ultimate failure of bioprosthetic heart valves.

Restoration of lateral hand grasp using natural sensors.

A closed-loop control system for controlling the key grip of a C6 tetraplegic patient was developed. Natural sensors served as the source of the feedback signal. The neural signals from cutaneous receptors were picked up by an implanted cuff electrode placed around the radial branch of the median nerve innervating the lateral part of the index finger. Mechanical stress applied to the skin, like pressure and slips, resulted in an increase in amplitude of the recorded neural signal. The goal of the study was to determine whether the recorded neural signals were able to indicate the slip of an object during lateral grasp and whether the slip could be stopped by increasing the grasp force through functional electrical stimulation of the thumb adductor and flexor.

Input-output nonlinearities and time delays increase tracking errors in hand grasp neuroprostheses.

Tracking tasks were designed to study how different input-output relationships (nonlinearities and time delays) would affect the performance of functional neuromuscular stimulation (FNS) hand grasp neuroprostheses. Simulated hand grasp neuroprostheses and real hand grasp neuroprostheses with and without closed-loop control were evaluated, with the subjects adjusting the input to the system so that the output would match a visual target track at three different bandwidths (able bodied subjects with the simulated systems and neuroprosthesis users with the real systems). For both systems, the tracking error increased as the input-output nonlinearity increased. Other factors that affected tracking performance were the target bandwidth and delays in the neuroprosthesis. The results support the hypothesis that hand grasp neuroprostheses with linear input-output properties and no delays will be controlled more accurately.