Pattern Recognition Based Prosthesis Control For Movement Of Forearms Using Surface And Intramuscular Emg Signals release_cacxnfsxorb27ghsm2doscufc4

Abstract

Myoelectric control system is the fundamental<br> component of modern prostheses, which uses the myoelectric signals<br> from an individual's muscles to control the prosthesis movements.<br> The surface electromyogram signal (sEMG) being noninvasive has<br> been used as an input to prostheses controllers for many years.<br> Recent technological advances has led to the development of<br> implantable myoelectric sensors which enable the internal<br> myoelectric signal (MES) to be used as input to these prostheses<br> controllers. The intramuscular measurement can provide focal<br> recordings from deep muscles of the forearm and independent signals<br> relatively free of crosstalk thus allowing for more independent<br> control sites. However, little work has been done to compare the two<br> inputs. In this paper we have compared the classification accuracy of<br> six pattern recognition based myoelectric controllers which use<br> surface myoelectric signals recorded using untargeted (symmetric)<br> surface electrode arrays to the same controllers with multichannel<br> intramuscular myolectric signals from targeted intramuscular<br> electrodes as inputs. There was no significant enhancement in the<br> classification accuracy as a result of using the intramuscular EMG<br> measurement technique when compared to the results acquired using<br> the surface EMG measurement technique. Impressive classification<br> accuracy (99%) could be achieved by optimally selecting only five<br> channels of surface EMG.
In text/plain format

Archived Files and Locations