Learning to operate a high-dimensional hand via a low-dimensional controller

Current status and clinical perspectives of extended reality for myoelectric prostheses: review

Training with "Extended Reality" or X-Reality (XR) systems can undoubtedly enhance the control of the myoelectric prostheses. However, there is no consensus on which factors improve the efficiency of skill transfer from virtual training to actual prosthesis abilities. This review examines the current status and clinical applications of XR in the field of myoelectric prosthesis training and analyses possible influences on skill migration. We have conducted a thorough search on databases in the field of prostheses using keywords such as extended reality, virtual reality and serious gaming. Our scoping review encompassed relevant applications, control methods, performance evaluation and assessment metrics. Our findings indicate that the implementation of XR technology for myoelectric rehabilitative training on prostheses provides considerable benefits. Additionally, there are numerous standardised methods available for evaluating training effectiveness. Recently, there has been a surge in the number of XR-based training tools for myoelectric prostheses, with an emphasis on user engagement and virtual training evaluation. Insufficient attention has been paid to significant limitations in the behaviour, functionality, and usage patterns of XR and myoelectric prostheses, potentially obstructing the transfer of skills and prospects for clinical application. Improvements are recommended in four critical areas: activities of daily living, training strategies, feedback, and the alignment of the virtual environment with the physical devices.

Autoencoder-based myoelectric controller for prosthetic hands

In the past, linear dimensionality-reduction techniques, such as Principal Component Analysis, have been used to simplify the myoelectric control of high-dimensional prosthetic hands. Nonetheless, their nonlinear counterparts, such as Autoencoders, have been shown to be more effective at compressing and reconstructing complex hand kinematics data. As a result, they have a potential of being a more accurate tool for prosthetic hand control. Here, we present a novel Autoencoder-based controller, in which the user is able to control a high-dimensional (17D) virtual hand via a low-dimensional (2D) space. We assess the efficacy of the controller via a validation experiment with four unimpaired participants. All the participants were able to significantly decrease the time it took for them to match a target gesture with a virtual hand to an average of 6.9 s and three out of four participants significantly improved path efficiency. Our results suggest that the Autoencoder-based controller has the potential to be used to manipulate high-dimensional hand systems via a myoelectric interface with a higher accuracy than PCA; however, more exploration needs to be done on the most effective ways of learning such a controller.