Crosstalk Removal in Stimulus-evoked Forearm Electromyography

Abstract

Abstract This paper presents the detection and reduction of crosstalk in stimulus-evoked electromyography (EMG) recorded from four forearm flexors: Flexor Carpi Radialis (FCR), Palmaris Longus (PL), Flexor Digitorum Superficialis (FDS) and the Flexor Carpi Ulnaris (FCU). Electrical stimulation was applied distal to the elbow, stimulating the medial and radial nerves singly. As the innervating nerve was stimulated, the M-wave was recorded from the target muscle and crosstalk from non-target muscles: for ulnar nerve stimulation, the M-wave was recorded from the FCU, and the FCR, PL and FDS were monitored for crosstalk. The branched electrode (BE) tripolar configuration was utilised for crosstalk reduction. The BE and double differential configurations are mathematically equivalent, only scaled by a factor of two. The benefit of the BE configuration is its ability for being used with a standard EMG device. A normalised peak-to-peak amplitude-based index was used to quantify the reduction in crosstalk, achieving a maximum reduction of 42.51% on the FDS, 12.43% on the PL and 7.68% on the FCU. An increase in peak-to-peak amplitude was recorded for the FCR, this is believed to be due to the challenge associated with electrode-skin impedance matching.