Numerical Simulations to Analyze and Optimize the Human Substitute Voice

After a larynx excision, as consequence of e.g. laryngeal cancer, the base of the human voice gets lost and a substitute voice is needed to communicate. A possible base of such a substitute voice represents the upper part of the esophagus, called the pharyngeal-esopohageal segment (PE segment). Via a valve, which the surgeon places between the esophagus and the trachea, the patient can guide air from the trachea into the esophagus. By virtue of that air stream the PE segment starts to vibrate and sound is generated, which can be used as base of a substitute voice. The geometry of this PE segment has major influence on the engaging voice quality. During surgery the surgeon is able to form the geometry of this PE segment. But until now the surgeon has no precise guidelines or knowledge of how to shape it in an optimal way. The target of this project is therefore to built a simulation tool which is capable to value different PE segment geometries in order to improve the quality of the substitute voice.

Within this contribution the mathematical models of the involved physical fields as well as their interactions will be discussed. The arising partial differential equations will be solved by using the finite element method (FEM). For the fluid field the weighted least-squares method (LSFEM) and for the mechanical field the Galerkin method is applied.

The numerical schemes are finally used to perform a geometry variation of the PE segment. Three different geometries of the PE segment opening (pseudo glottis) -a round, an elliptic and a triangularare compared with each other in respect to their dynamical behavior under fluid loads.