Surface Electromyographic Activity in Total Laryngectomy Patients following Laryngeal Nerve Transfer to Neck Strap Muscles

Abstract

Hands-free triggering and pitch control would improve electrolarynx devices, which are inconvenient to use and sound unnatural. The present study tested the strategy of salvaging voice-related neural signals for hands-free electrolarynx control either by transferring cut recurrent laryngeal nerves (RLNs) to denervated neck strap muscles or by preserving strap muscles with their normal innervation. An RLN nerve transfer was performed at the time of total laryngectomy in 8 individuals, and in 5 of these subjects, strap muscles with intact ansa cervicalis innervation were also preserved for comparison. Neck surface electromyography performed over the strap muscles was used periodically for more than 1 year on phonatory and nonphonatory tasks. Signals were eventually obtained in all subjects from both RLN-innervated and ansa-innervated strap muscles that correlated with speech production. After 1 year, RLN-driven signals were larger than ansa-driven signals in magnitude, and their timing appeared better correlated with intended phonation. The results show that neck surface electromyography is an effective control source for hands-free electrolarynx activation, and that RLN transfer may provide the best approach for obtaining phonation-related activity.

Tonal Perceptions in Normal Laryngeal, Esophageal, and Electrolaryngeal Speech of Mandarin

“…However, the poor control of pitch in EL speakers is expected to change in the near future, as a new generation of electrolarynges is being designed and tested. The neural-controlled electrolarynx will allow the EL speakers of a tone language to manipulate the change of pitch by using the surviving laryngeal nerves instead of the fi nger [34] .…”

Vocal frequency estimation and voicing state prediction with surface EMG pattern recognition

“…In addition, while Heaton et al [30] indicate the existence of a consistent relationship between the intended pitch and EMG amplitude for individuals, there is significant variability across different subjects. This high variability is largely due to inherent physiological differences (e.g.…”

“…The F0 prediction for the user-independent model does perform above the chance threshold of 87.39, but not by much. Quite likely, the poor performance is a result of the significant intersubject variability in the EMG-F0 relationship reported in the literature [30,28]. This high variability is largely due to inherent physiological differences (e.g.…”

Normalization Patterns of the Surface Electromyographic Signal in the Phonation Evaluation

“…Accepting this argument, the MVSA can then be the chosen pattern. Although we evaluated both IHs, it is worth considering as other studies, the possibility to evaluate the muscle group unilaterally, when symmetry is not the objective of the study, 11,12,14,28 because phonation is a dynamic process of muscular management.…”