Prosthetic feedback systems

To fully replace the missing limb, a myoelectric prosthesis needs to provide a bidirectional communication between user's brain and its bionic limb. And indeed, modern prosthetic hands are advanced mechatronic systems that approach the design and capabilities of biological hands both morphologically (size, shape, and weight) and functionally (degrees of freedom). In addition, these hands are controlled intuitively by mapping muscles' activations to prosthesis functions using direct control or pattern classification. However, commercial systems do not yet provide somatosensory feedback to their users. In this chapter, we provide an overview of the methods and techniques that can be used to stimulate the sensory motor structures of an amputee subject in order to restore the missing sensations. We then discuss the prosthesis variables that are most often transmitted through the stimulation as well as the encoding schemes that can be used to map those variables to stimulation parameters. The contradictory evidence about the impact of feedback on the prosthesis performance is presented next, illustrating that designing, implementing, and assessing effective feedback interfaces is indeed a challenging task. Finally, the chapter ends with discussion and recommendation for further research that will hopefully lead to a successful solution for closed-loop prosthesis control.