| Popis: |
Bidirectional hand neuroprostheses for amputee subjects aim to restore sensory–motor functions via the decoding of motor intentions in efferent channels and the stimulation of afferent nerves with electrical pulses conveying sensory information and closing the control loop. To optimize sensory feedback, neural stimulation should evoke both informative and plausible sensations. This latter factor affects the level of embodiment, which is crucial to avoid prosthesis rejection, as plausible and naturalistic sensations reduce the patient’s discomfort and require less cognitive load. It also affects the steepness of the learning curve as it is more straightforward to associate the stimulation-induced perception with an intact hand. The amount of information conveyed by a given neural stimulation pattern and the similarity of the sensation with the natural ones depend on the choice of the encoding strategy, and there is, at the moment, no consensus on the optimal approach. For example, important elements of sensory feedback can be conveyed simply by modulation of the amplitude of the injected pulses; however, a natural temporal pattern of pulses for the encoded somatosensory feedback leads to more intuitive sensations. The challenges for the motor and bidirectional sensory–motor restoration are similar. Ideally, a seamless and intuitive control of the hand neuroprostheses using simultaneous and proportional decoding of the intended degrees of freedom is the aim. In this chapter, we describe the technical approaches and recent results for both biomimetic encodings of sensory feedback and the decoding of motor functions. |
