Online mapping of EMG signals into kinematics by autoencoding
- PDF / 1,332,664 Bytes
- 9 Pages / 595.276 x 790.866 pts Page_size
- 24 Downloads / 218 Views
RESEARCH
Open Access
Online mapping of EMG signals into kinematics by autoencoding Ivan Vujaklija1, Vahid Shalchyan2, Ernest N. Kamavuako3, Ning Jiang4, Hamid R. Marateb5 and Dario Farina1*
Abstract Background: In this paper, we propose a nonlinear minimally supervised method based on autoencoding (AEN) of EMG for myocontrol. The proposed method was tested against the state-of-the-art (SOA) control scheme using a Fitts’ law approach. Methods: Seven able-bodied subjects performed a series of target acquisition myoelectric control tasks using the AEN and SOA algorithms for controlling two degrees-of-freedom (radial/ulnar deviation and flexion/extension of the wrist), and their online performance was characterized by six metrics. Results: Both methods allowed a completion rate close to 100%, however AEN outperformed SOA for all other performance metrics, e.g. it allowed to perform the tasks on average in half the time with respect to SOA. Moreover, the amount of information transferred by the proposed method in bit/s was nearly twice the throughput of SOA. Conclusions: These results show that autoencoders can map EMG signals into kinematics with the potential of providing intuitive and dexterous control of artificial limbs for amputees. Keywords: Prosthetic control, Myoelectric signal processing, Regression, Online performance, Autoencoding
Background Myoelectric signals (EMG) have been used to drive prosthetic devices for more than half a century. However, the commercially available products still mainly rely on a simple direct and sequential control. This control strategy offers robust and reliable handling of the prosthetic in daily life, but it allows limited recovery of functionality and requires high cognitive load by the user [1, 2]. Therefore, several attempts have been made for establishing a more intuitive interface for active prosthesis control. Major advances in myocontrol have been made with pattern recognition approaches. These methods are based on the assumption that sufficiently distinguishable patterns can be observed in the EMG recordings during different motions. Each signal can be represented using a certain set of features which can be used as input to a classifier. The trained classifier is then capable of discriminating the intended motions. With state of the art pattern recognition methods, the classification accuracy exceeds > 95% when discriminating > 10 classes [3]. * Correspondence: [email protected] 1 Department of Bioengineering, Imperial College London, London, UK Full list of author information is available at the end of the article
Despite their good performance and their recent translation in commercial systems [4], pattern recognition algorithms for myocontrol have some intrinsic limitations. For example, it is difficult to implement simultaneous and proportional control of multiple degrees of freedom (DoFs) with these algorithms since they do not allow a direct mapping of EMG into kinematics. This issue can be mitigated by the classification of complex movements as the co
Data Loading...
